S. Hrg. 103-457
EVALUATION OF THE U.S. STRATEGIC NUCLEAR
TRIAD
Y 4. G 74/9: S. HRG. 103-457
Evaluation of the U.S. Strategic Ku. .
HEARING
BEFORE THE
COMMITTEE ON
GOVERNMENTAL AFFAIRS
UNITED STATES SENATE
ONE HUNDRED THIRD CONGRESS
FIRST SESSION
JUNE 10, 1993
Printed for the use of the Committee on Governmental Affairs
may 2 ^
U.S. GOVERNMENT PRINTING OFFICE
69-539 cc WASHINGTON : 1994
For sale by the U.S. Government Printing Office
Superintendent of Documents, Congressional Sales Office, Washington, DC 20402
ISBN 0-16-043971-X
S. Hrg. 103-457
EVALUATION OF THE U.S. STRATEGIC NUCLEAR
TRIAD
Y 4. G 74/9: S. HRG. 103-457
Evaluation of the U.S. Strategic Nu. .
HEARING
BEFORE THE
COMMITTEE ON
GOVERNMENTAL AFFAIRS
UNITED STATES SENATE
ONE HUNDRED THIRD CONGRESS
FIRST SESSION
JUNE 10, 1993
Printed for the use of the Committee on Governmental Affairs
may 2$
*-2l,
U.S. GOVERNMENT PRINTING OFFICE
69-539 cc WASHINGTON : 1994
For sale by the U.S. Government Printing Office
Superintendent of Documents, Congressional Sales Office, Washington, DC 20402
ISBN 0-16-043971-X
COMMITTEE ON GOVERNMENTAL AFFAIRS
JOHN GLENN, Ohio, Chairman
SAM NUNN, Georgia WILLIAM V. ROTH, Jr., Delaware
CARL LEVIN, Michigan TED STEVENS, Alaska
JIM SASSER, Tennessee WILLIAM S. COHEN, Maine
DAVID PRYOR, Arkansas THAD COCHRAN, Mississippi
JOSEPH I. LIEBERMAN, Connecticut JOHN McCAIN, Arizona
DANIEL K. AKAKA, Hawaii
BYRON L. DORGAN, North Dakota
Leonard Weiss, Staff Director
David Hafemeister, Professional Staff Member
Franklin G. Polk, Minority Staff Director and Chief Counsel
Michal Sue Prosser, Chief Clerk
(ID
CONTENTS
Opening statements: Pagf
Senator Glenn *
Senator Cohen
Prepared statement:
Senator Dorgan
WITNESSES
Thursday, June 10, 1993
Eleanor Chelimsky, Assistant Comptroller General for Program Evaluation
and Methodology, U.S. General Accounting Office ...■.■■■ 4
William J. Perry, Deputy Secretary of Defense, U.S. Department of Defense ... Id
Alphabetical List of Witnesses
Chelimsky, Eleanor:
Testimony 4
Prepared statement dy
Perry, William J.: n _
Testimony £>
Prepared statement 4»
APPENDLX
Prepared statements of witnesses in order of appearance 39
Team B: The Trillion j>2
Dollar Experiment °'
Senator Glenn news release dated June 28, 1993 od
Letter, with enclosures, dated June 16, 1993, to Senator Glenn from Eleanor
Chelimsky, Assistant Comptroller General, GAO 64
Letter, dated June 28, 1993 to Secretary Les Aspin from Senator Glenn 88
Letter, dated June 28, 1993 to Mr. Vander Schaaf from Senator Glenn 91
Letter with enclosure, dated August 6, 1993 to Senator Glenn from Eleanor
Chelimsky, Assistant Comptroller General, GAO 94
Written questions submitted by Senator Dorgan and GAO responses 102
Written questions submitted by Senator Cochran and GAO responses 106
Inserts for the Record — responses to questions from Senators Glenn and
Cochran:
DOD Analysis on the Triad llj*
The Clinton-Aspin Nuclear Force 11?
Uncertain Costs jl°
B-52 vs. Trident>— Cost Per Surviving Warhead 119
ICBMs vs. SLBMs J21
Minuteman III Modernization 122
B-52 Lifetime 124
Triad Cost-Effectiveness for the Future 125
DOD Position on GAO Evaluation 126
Triad Support 127
Resumption of Underground Testing 12o
(III)
EVALUATION OF THE U.S. STRATEGIC
NUCLEAR TRIAD
THURSDAY, JUNE 10, 1993
U.S. Senate,
Committee on Governmental Affairs,
Washington, DC.
The Committee met, pursuant to notice, at 9:35 a.m., in room
SD-342, Dirksen Senate Office Building, Hon. John Glenn, Chair-
man of the Committee, presiding.
Present: Senators Glenn, Levin, Lieberman, Dorgan, and Cohen.
OPENING STATEMENT OF CHAIRMAN GLENN
Chairman Glenn. The hearing will be in order.
We are going to change our usual format of the Committee this
morning because we have some time constraints on the second wit-
ness this morning. Bill Perry, Deputy Secretary of Defense, will ar-
rive shortly. He has other commitments this morning, but he
worked his schedule so he could be with us for 1 hour, from 10 to
11. So we will go ahead and have the Hon. Eleanor Chelimsky, As-
sistant Comptroller General for Program Evaluation and Methodol-
ogy of GAO, give her testimony, and answer questions until 10
o'clock, and then we will go ahead with Bill Perry's testimony. Fol-
lowing his testimony, we shall then return to Ms. Chelimsky for
more questions.
I will enter my opening statement in the record and just open
with a few remarks.
Throughout the Cold War period, our nuclear deterrent has con-
sisted of a triad of land-based, sea-based, and air-based ability to
deliver nuclear weapons.
There has been a lot of comment about the triad through the
years, including some of my own, because I have felt that perhaps
in some aspects we did not need all of the traid. I thought that if
we got into a nuclear war and we had ICBM's and SLBM's and
cruise missiles all delivering nuclear weapons, that to ask our peo-
ple to get in an airplane and go in and add one or two more bombs
to what was already going off was too much.
I have supported the bombers because I felt that in war we
would need a conventional war capability and I have not supported
the bombers on their nuclear delivery capability. We got the nu-
clear capability as a freebie. That was my view on this for a long
time.
The GAO, in response to a House Foreign Affairs request, started
a study on the triad two years ago. It has been a very major study,
I think one of the biggest ones GAO has ever done. Mr. Bowsher
(1)
told us that document after document after document are still clas-
sified secret, which we will obviously not release today. The GAO
has gone into all sorts of detail on the triad. They have done a very
thorough study of it, and that is why we are here today. We will
examine the process by which those decisions were made during
the days of the Cold War, so we can learn something from it.
This is not supposed to be just a critical hearing to point fingers
at anybody. It is supposed to let us learn how we made those deci-
sions, so that as we are making decisions on other weapons sys-
tems now, which we will be doing, we can do it on a more rational
basis in order to avoid some of the mistakes that we made during
that period of time.
Mr. Perry will be along at 10 and be prepared to answer some
questions.
Prepared Statement of Senator Glenn
Throughout the Cold War era, America's security against nuclear attack was
maintained by a defense posture based on the threat of massive nuclear retaliation
against any such attack. For some 40 years, this capability was maintained using
long-range bombers, intercontinental ballistic missiles (ICBMs) and submarine-
launched ballistic missiles (SLBMs). These three elements comprised the subject of
today's hearing — namely the evaluation of the U.S. strategic triad.
In 1990, Chairman Dante Fascell of the House Foreign Affairs Committee re-
quested the General Accounting Office to evaluate the strengths.and weaknesses of
the triad and determine which triad upgrades appear to be most cost-effective. On
September 28, 1992, the GAO released its unclassified results, which raise a num-
ber of serious questions not yet addressed by Congress, which will be the basis for
today's hearing. In addition, GAO released eight classified reports on the many de-
tails of the study. It is the Governmental Affairs Committee's role to examine the
"efficiency and effectiveness" of the U.S. government, and it is in that spirit that
we meet here today.
We can all thank our lucky stars that the Cold War ended with a peaceful wimper
and not a catastrophic bang. Some believe that the Cold War ended on November
19, 1990 with the signing of the CFE Treaty in Paris when the Soviets gave up a
3 to 1 advantage in conventional forces. This reduction in conventional weapons led
the way to reductions in strategic nuclear weapons in the START treaties. On that
day in November 1990, President Mitterrand of France wisely reminded us that
there were "neither victors nor vanquished." Both sides sacrificed a great deal in
this contest, but at least direct confrontation was avoided.
Nonetheless, because of continued uncertainties, we and the Russians will con-
tinue to have nuclear strategic weapons for some time to come. Because of this like-
lihood, Congress must examine our past experience with the triad to find lessons
that can guide us to a better future. We can estimate the Cold War arms build-up
cost at some $2 trillion, and the cost of the triad, including DOE clean-up, at more
than one-half trillion. Peace and freedom are worth a very high price, but it is also
true that we could have saved billions if we had been able to predict the future with
certainty and better data.
A key conclusion of the GAO report is that "there exist systematic disparities be-
tween estimates or claims that have been made about the triad systems and what
the data actually show. We found this to be the case whether the issue was the like-
ly cost and performance of upgrades, the actual performance of current systems, or
the likely offensive or defensive threats to these systems from the former Soviet
Union." This is a serious assertion — in short, all the errors on actual and predicted
performance and on predicted threats pointed to the same conclusion, that of foster-
ing additional procurements for the strategic forces.
In addition, GAO states that their study on the triad will be useful to Congress
for making decisions on future strategic forces "because it is the first study in at
least three decades that sets up a comprehensive framework for comparing numer-
ous dissimilar strategic systems on multiple measures and that uses test and per-
formance data to compare the systems in question." If this claim is true, then one
can only wonder how so much could be spent without such a study ever having been
performed.
Today's hearing will dwell mostly on process, to try and answer the question "how
can we spend our declining financial resources more wisely?" However, I would ex-
pect that our witnesses will also be asked questions on specific triad systems.
It is my distinct honor to welcome our two witnesses who will address the issues
raised in the GAO report on the triad. Our first witness will be the Honorable Elea-
nor Chelimsky, Assistant Comptroller General for Program Evaluation and Meth-
odology, United States General Accounting Office. Our second witness on a separate
panel will be the Honorable William J. Perry, Deputy Secretary of Defense, Depart-
ment of Defense.
Chairman Glenn. Senator Cohen.
OPENING STATEMENT OF SENATOR COHEN
Senator COHEN. Thank you, Mr. Chairman, first of all for rec-
ognizing me today. I know you have great difficulty looking to your
right these days. [Laughter.]
Chairman Glenn. I am being needled for sitting here yesterday,
reading my statement and not recognizing Senator Cohen who had
walked in. I turned and asked another Senator to speak in im-
proper order. I have been chastised for this all day yesterday.
Senator Cohen. Thank you, Mr. Chairman.
I don't have an opening statement, but the debate that we have
had over the years is whether or not we should first devise a strat-
egy and then appropriate the dollars to match the strategy, rather
than looking at budgetary constraints and then shaping our strat-
egy to match the budget numbers.
That is a debate, I suspect, that will even continue here today.
We know, for example, that Gen. Powell had a base force concept.
That concept, it seems to me, under the proposal coming from the
administration, has been pretty much discarded. Instead of having
a two regional war type of strategy, it seems that the Secretary of
Defense has come up with a so-called win-hold-win strategy, and
it may be portrayed as an effort to close the gap between our capa-
bilities and our strategy, but I would suggest that it might have
an adverse impact upon our strategic interest in the future.
I won't take the time until Secretary Perry arrives to explore this
with him, but I will do so at that time.
Mr. Chairman, I suggest that this win-hold- win strategy really is
simply a rationalization, in the final analysis of the budgetary con-
straints that we now find ourselves faced with. We see, for exam-
ple, that Secretary Aspin has already proposed real cuts in defense
budget authority of 17 percent between fiscal 1993 and fiscal 1997,
compared to former President Bush's planned cut of 8 percent oyer
that same period. So I think these cuts, along with the uncertain-
ties in the Secretary's and in the President's plan really mean we
cannot afford Gen. Powell's base force concept. But I will talk about
that more when Secretary Perry comes.
Chairman Glenn. Thank you very much.
Ms. Chelimsky, welcome to our hearing this morning. Please pro-
ceed with your testimony.
TESTIMONY OF ELEANOR CHELIMSKY,* ASSISTANT COMP-
TROLLER GENERAL FOR PROGRAM EVALUATION AND
METHODOLOGY, U.S. GENERAL ACCOUNTING OFFICE
Ms. Chelimsky. Thank you so much, Mr. Chairman. Good morn-
ing. Good morning, Senator Cohen. Your argument reminded me of
that phrase of Muriel Sparks about Job, that he had suffered the
problem of argument, but he also argued the problem of suffering.
It is a great pleasure to be here today — excuse my voice, I have
one of these terrible colds that never goes away — to talk about
GAO's evaluation of the strategic triad. I really want to thank you,
Mr. Chairman, for holding this hearing on the study. We appre-
ciate it.
Let me begin by introducing Kwai Chan, who directed this work,
as well as Winslow Wheeler, who is in back of me and who is our
Assistant Director for Defense Studies; and Jonathan Tumin, our
Project Manager.
Abraham Lincoln said that statements should be short and
sweet, like the old woman's dance, and I really subscribe to that,
except with this voice, I think it will be short and hoarse, but I will
try to be as short as I possibly can, given the eight studies that
we have produced addressing policy questions like how vulnerable
are our submarines and our silo-based ICBM's? How effective are
our ballistic missiles that are land launched, compared to ballistic
missiles that are sea-launched? How good are the air legs' proposed
upgrades relative to the systems they are going to replace? So I
will try to be short and sweet.
Let me make just three points about the scope, the methodology
and the origins of our evaluation, before getting into the findings.
First, what does the study include? Well, the nuclear weapon sys-
tems and proposed upgrades we included in the evaluation were for
the air leg: The B-52G and H, B-1B and B-2B bombers, as well
as the ALCM, ACM, SRAM A and SRAM II missiles.
For the land leg, we looked at Minuteman II and Minuteman III
ICBM's, the Peacekeeper Rail Garrison and the small ICBM. For
the sea leg, we examined the C-4 and D-5 SLBM's on Lafayette
and Ohio-class nuclear-powered ballistic missile subs. We assessed
all systems under a full range of threat scenarios, moving from
total surprise attack to strategic warning.
Second, what was our methodology? Well, our basic approach
was to develop a framework for comparison. Because we found no
earlier comparative studies on which to build, we had to develop
our own set of measurements. This involved using the same 30-
year life-cycle cost methodology to compare upgrade costs, and the
same measures of effectiveness to compare weapon system perform-
ance. Our method was to examine the Defense Department's con-
clusions about the performance of the various triad weapons sys-
tems, the costs of the upgrades being proposed and the size and na-
ture of the Soviet threat. We then looked for the qualitative and
quantitative evidence that was needed to support and validate the
DOD conclusions.
We collected our quantitative data through an exhaustive review
of the technical literature, including all the relevant DOD tests and
1 The prepared statement of Ms. Chelimsky appears on page 39.
evaluations that we could get. The data we drew on came from a
wide variety and large quantity of classified data sources, about
250 major technical reports in all. These included 52 for the land
leg alone, 71 for the air leg, 33 for the sea leg and 43 for the Soviet
threat.
We collected our qualitative data through interviews. These also
were extensive. We visited field sites, military commands and
bases, as well as program offices. In addition to the special 14-
member very prestigious advisory board that we constituted for
this study, we consulted military and civilian experts in a range of
agencies, universities and think tanks. Our interviews took us to
the Joint Chiefs of Staff, the CIA, the DIA, DARPA, the Center for
Naval Analyses, the Institute for Defense Analysis, the MITRE
Corporation, the Rand Corporation, ANSER, Johns Hopkins Ap-
plied Physics Lab, the Arms Control and Disarmament Agency,
CBO, CRS, OTA, many specialized DOD bureaus— for instance, the
Defense Mapping and Nuclear Agencies — as well as the relevant
service and OSD offices, plus experts at Brookings, MIT and a host
of other universities.
Senator Cohen. Anybody in Congress?
Ms. Chelimsky. A few people in Congress, yes. We never men-
tion them.
In all, we did more than 200 major interviews. This was not a
simple study, either conceptually, in its process, or logistically. We
studied performance in weapons systems, across weapons systems
and across legs, and we looked carefully at synergy along four di-
mensions, in what way would they complement enemy defensive
and offensive strategy planning, how would it dilute enemy re-
sources, providing technical hedges against potential enemy break-
throughs and providing flexibility response options to the U.S.
President. So we did not look at just cost and performance alone.
My third point is on the study's origins. As you have mentioned,
Mr. Chairman, it was the House Foreign Affairs Committee that
asked us to address the major strategic modernization programs of
the Carter, Reagan and Bush administrations. In particular, what
they wanted was a databased evaluation to establish the strengths
and weaknesses of these programs and the determination of which
among them appeared to be the most cost-effective.
These eight reports, then, constitute what I would call a baseline
evaluation of the cost-effectiveness of this $350 billion moderniza-
tion program. More evaluations need to be done.
Now let me turn to our findings, first of all those regarding
weapon systems. Trying again to keep this short and sweet, I will
report only seven of these, but there are many, many more.
Comparing across the triad legs, we found that, on balance, the
sea leg emerges as the most cost-effective of the legs. What is the
basis for this finding? Well, let me mention just three points.
First, both the speed and the reliability of day-to-day communica-
tions to submerged deployed SSBN's are far better than widely be-
lieved, and about equal in speed and reliability of communications
to ICBM silos. Contrary to what we expected, SSBN's are in essen-
tially constant communication with national command authorities
and, depending on the scenario, SLBM's from nuclear submarines
would be almost as prompt as ICBM's in hitting enemy targets.
Second, we found that the accuracy of the D-5 is about equal to
that of the Peacekeeper. Its reliability also is about equal. Its war-
head had a higher yield than the Peacekeeper's, and we estimate
that the D-5 has a hard target kill capability about equal to the
Peacekeeper. Further, unlike easily located silos, the submerged
SSBN's are essentially invulnerable. They are even less detectable
than generally understood, and there appear to be no current or
long-term technologies that would change this.
Third, cost comparisons also turned out to be favorable. Meas-
ured against the B-2, for example, in terms of life-cycle costs for
arriving warhead, we found that the B-2 would cost two and a half
to five times more than the D-5 under any attack scenario, depend-
ing on the number of warheads on the D-5.
In addition, these estimates favor the B-2, because they assume
the bomber will be as effective as DOD plans it to be, and that
costs will not grow. Whereas, for the D-5 Ohio system, we already
have considerably more reliable and complete data on cost and
operational performance.
When we compared the de-MIRVed Minuteman III (now being
proposed by the Air Force) to the D-5 Ohio, using life-cycle cost-
to-go per arriving warhead, we had found that the cost for the two
systems was approximately the same, but that was based on an Air
Force error. Now, corrected life-cycle cost figures for the Minute-
man III increase it from $16 to $23 billion, which makes the D-
5 Ohio system the less costly of the two.
Senator Cohen. Do you have to call it the Ohio system?
Ms. Chelimsky. I'm sorry?
Senator Cohen. Do you have to call it the D-5 Ohio system?
Chairman Glenn. I think it is a nice-sounding name, myself.
[Laughter.]
Ms. Chelimsky. In short, it is these favorable outcomes that
Chairman Glenn. I can't imagine naming a submarine the
Maine system, can you? [Laughter.]
Ms. Chelimsky. I was going to leave that one entirely alone, but
I see I am not going to be allowed to do that.
In short, it is these favorable outcomes, all of them, with respect
to communications, cost, accuracy, reliability, warhead yield, hard-
target-kill capability and, most importantly, invulnerability, that
led to our finding on the sea leg's cost-effectiveness. I wanted to
make that abundantly clear. It is not just an issue of cost. There
are many, many issues that we looked at.
A second finding deals with the uncertainties in the B-2 pro-
gram, in particular, the aircraft's stealth characteristics. We found,
as did the Air Force, a number of performance weaknesses that
seemed the more problematic because only about one-quarter of the
bomber's flight-testing hours have been flown. Since important and
expensive problems can show up either late in the testing process
or after deployment, as we have all seen, this means the jury is
still out on B-2 performance. As for the B-2's cost, we found that
each B-2 for a 20-aircraft fleet will cost over $2 billion in develop-
ment and procurement, and there will also be added operating
costs when the program is complete. Yet, the five additional B-2's
requested by DOD will increase total air leg strategic warheads by
only 2.3 percent above current numbers.
A third finding concerns the B-1B. Although this aircraft is less
technologically ambitious than the B-2, its own problems with
stealth capability have not been resolved. Further, we found DOD's
presentation of these problems to be misleading, especially with re-
gard to conclusions about the B-lB's radar cross-section. Finally,
a long history of recognized test and operational shortcomings in
the aircraft, especially its electronic countermeasures, flight con-
trols, range and reliability, raise questions about both feasibility
and cost.
Our fourth finding was as unexpected to us as reliability and
speed of communications to SSBN's, namely, the continued viabil-
ity of the B-52, whose obsolescence has been so widely reported
and so often cited as a rationale for procuring both the B-1B and
the B-2. Instead, the data show that both the B-52G's and H's will
remain usable aircraft for years to come. There are three reasons
for this.
First, Air Force flight-hour data show that, as of 1990, the air-
frames and other key structural components of both models had
reached only about half their life expectancies. Indeed, an Air Force
analysis on the effects of aging in the B-52 addressed multiple age-
related factors and found that both the B-52G's and the B-52H's
would remain structurally sound until the year 2030, if they were
flown at the current average number of flight-hours per year.
The second reason is that the Air Force has continued to modify
the B-52 to assure its continuing effectiveness, has indeed done an
admirable job with regard to that, with new ECM, new passive sen-
sors, new communications equipment, new navigation gear and
new weapons, both conventional and nuclear. So this is a system
that has been kept up.
Finally, comparisons of data on multiple measures of effective-
ness show that the B-52 compares favorably to the newer B-1B on
a number of important performance dimensions. Both models of the
B-52 have continuing capability, the B-52G as a cruise missile car-
rier, the B-52H as a strategic penetrating bomber.
A further rationale for procuring the B-2, other than the B-52's
alleged obsolescence, was a projected increase in the Soviet air de-
fense threat. Our fifth finding is that this increase never took
place. The data show, and interviews confirm, that Soviet air de-
fenses were considerably lower than the number forecast earlier,
and that their effectiveness had also been greatly overstated.
Today, conditions in the former Soviet Union are such that cur-
rent air defenses are much more likely to degrade than improve.
In other words, the air defense threat that the B-2 purported to
address was never in fact deployed, and this can be seen before
Gorbachev came to power, quite a bit before.
Our sixth finding involves the predicted benefits of the ACM ver-
sus the ALCM. Here we found that the range requirement for the
ACM provided only a small improvement over what the older
ALCM already could do, and that the accuracy improvement of-
fered also doesn't appear to have much real operational signifi-
cance. Further, because of important limitations in the flight tests
conducted for both cruise missile systems, their performance under
operational conditions remains uncertain.
8
Our seventh finding, the last one I am going to report here on
weapon systems, deals with our silo-based ICBM's and the expecta-
tion of their increased vulnerability. We found this questionable on
at least two counts. First, the expectation was premised on worst
case calculations, that is, assuming only the highest estimates for
Soviet missile accuracy, yield and reliability, while at the same
time discounting substantial performance uncertainties that could
only have been resolved in a nuclear war.
Second, it ignored the deterrence function of the triad's other
legs. This is the issue of synergy, which we talk about so often as
something that we absolutely must have, except when the time
comes to rationalize weapons systems. Then often look at them
without considering the potential of the other legs to deal with
some of the problems a given leg may have. What was the purpose
of having the triad, if not to dissuade the Soviets from an all-out
attack on one leg, through fear of retaliation by the others? In this
case, of course, I mean retaliation by submerged submarines or on-
alert bombers and their thousands of warheads.
Now let me turn to our findings on evaluation. A general conclu-
sion from our study is that there exist systematic disparities be-
tween what the data showed and DOD's claims and estimates for
(1) the Soviet threat, (2) the performance of mature systems, and
(3) the expected performance and costs of proposed upgrades. I say
systematic disparities, because they seem to follow a particular pat-
tern, tending to overstate threats to our weapon systems, to under-
state the performance of mature systems, to overstate the expected
performance of upgrades, and to understate the expected costs of
those upgrades.
Specifically, we found that the vulnerability of our B-52's, subs,
and silo-based ICBM's to a Soviet threat had been overstated; that
performance claimed for the B-2's is yet unproven; that B-1B ACM
and Peacekeeper capabilities were often inflated; that costs for
strategic systems generally were incomplete (operating and support
costs having typically gone unreported), and that the performance
of B-52's and SSBN's were consistently understated.
The two charts that you see over there summarize the different
kinds of problems we found. The one on the right deals with knowl-
edge limitations based on insufficient or inadequate testing, and
with reporting problems in which conclusions don't necessarily flow
from the data in hand. The chart on the left gives specific examples
of these reporting problems in which we were surprised to find that
the data did not support six widely-held beliefs or conclusions on
triad characteristics.
In sum, and with a few happy exceptions, we were disappointed
by the number of problems we found in the quality and the objec-
tivity of testing, forecasting and reporting. More importantly, how-
ever, we were troubled by weaknesses in performance measure-
ment and program accountability arising from the lack of compara-
tive evaluation.
Our recommendations to the Congress are six in number. First,
we see no requirement for five more B-2's. Second, we believe the
B-1B needs a lot more operational testing to verify that its long-
standing performance problems have been resolved.
Third, given the estimated $23 billion cost of upgrading and
maintaining the de-MIRVed Minuteman III force through the year
2020, given also today's low-threat environment, we question the
advisability of funding major life-service upgrades for this force.
Fourth, we urge that D-5 testing be continued at adequate lev-
els, to insure confidence in the missile's performance, continuing
confidence.
Fifth, we agree with DOD's decision on the ACM, but believe
more testing is needed for both the ACM and the ALCM.
Last, we would encourage the Congress, and especially this Com-
mittee, with its concern for objective, accurate and accountable per-
formance measurement, to request two kinds of comparative eval-
uations from DOD. First, those that examine whether proposed up-
grades will really add capability to existing systems, and, if so, at
what cost. And, second, those that compare the performance and
costs of like weapon systems across the triad legs. Performing these
types of studies carefully and objectively should greatly improve
current uncertainties with regard to weapon system performance
and cost.
That concludes my remarks, Mr. Chairman, and I would be
pleased to answer any questions.
Chairman Glenn. Thank you very much, Ms. Chelimsky.
Secretary Perry is here and we will get to him in just a moment.
I just have one question now. What were the limitations that were
put on this GAO study?
Ms. Chelimsky. Yes, there are always limitations on every
study, and this one has them, too.
Chairman Glenn. What were they, so we will know what the
basis was?
Ms. Chelimsky. I feel that the major limitation — and there are
others I am sure that other people will point out — it is a first study
and, of course, it has probably more limitations than others, but I
think the chief limitation is the dependence on DOD data and the
problems we found with the DOD data, because our study depends
entirely on the data that were given us.
We had problems with missing data, data that were so sketchy
that we could not really validate conclusions that had been made.
In those cases, we always gave DOD the benefit of the doubt. We
mentioned that there was an uncertainty, but we used the data
that DOD gave us, even if it was not validated.
The second problem we had with the data was exclusion, often
through classification. In one case we were denied data that we
really thought that we should have, and Mr. Bowsher sent to Sec-
retary Cheney a demand letter for it. This was on the Peacekeeper
warhead reliability data. That was never ever given to us, which
was a very important problem.
I think we were also given misleading data.
Chairman Glenn. Misleading data?
Ms. Chelimsky. Misleading data. The $16 billion for the Minute-
man that I was talking about earlier, we were told that it included
costs to the year 2020. Instead they went only to the year 2010.
And only after we had published our reports did they rectify the
error. I am sure this was an error, but it was misleading: we had
to redo the calculations because of that, and it made a big dif-
10
ference in our comparisons between the Minuteman and the D-5
Ohio.
Finally, the most annoying, we were given some falsified data,
and this was on the B-1B radar cross-section. We have a memo in
our possession that shows what data GAO should have received
and what were the real data. Of course, that was a problem for us
and we went back, of course, and we got that ironed out rather rap-
idly, but it did
Chairman Glenn. The memo indicated that false data was to be
given to you. Was that it?
Ms. Chelimsky. It said what data should be given to the GAO
and what data they were giving to somebody else. I don't know how
we got this memo. It was sent to us by accident, clearly, but we
have the memo, so we knew that the data we had on the B-1B
radar cross-section was false.
Chairman Glenn. I want to ask more questions on this later.
Senator Cohen, do you have any questions before we go on?
Senator Cohen. One of the questions I was going to ask in terms
of in what way it was misleading, and I think you have already
clarified. It was not misleading, it was
Ms. Chelimsky. I would like to end with one hopeful point, and
that is that with Secretary Aspin and Secretary Perry here and
their review of the bottom-up data, we are hoping that other data
are going to be produced. In fact, I have no doubt about it. But we
didn't find a very good database.
Chairman Glenn. Ms. Chelimsky will still be here for questions
after Mr. Perry has to leave.
Senator Cohen. Could I just ask one quick question?
Chairman Glenn. Certainly.
Senator Cohen. If you were not given all of the information, if
some of the more highly classified information was denied to you,
can you still have reliability or can we have reliability in some of
your conclusions? For example, you were very clear on the surviv-
ability of the Ohio, which may or may not be survivable in other
contexts. But at least with respect to the D-5 sub, there was infor-
mation within the intelligence community — and I am sure Senator
Glenn, having a number of
Ms. Chelimsky. Yes, we have been briefed on it and we are very
aware. In that case, we had no problem. We were given all the data
that we needed. We only had one basic problem, I think, that was
of serious concern to us, and that was the Peacekeeper warhead re-
liability data.
Senator COHEN. But you had everything, as far as non-aperture
radar and other types of systems?
Ms. Chelimsky. Right. There were many things that we could
not write about, because they were too highly classified.
Senator Cohen. But you had access?
Ms. Chelimsky. Yes, we had access.
Senator Cohen. Thank you.
Chairman Glenn. Senator Dorgan, do you have any questions
before we move on?
Senator Dorgan. I do have questions, but if she will remain, I
will be happy to give her the
Ms. Chelimsky. I will.
11
Chairman Glenn. If you will stand by, we would appreciate that.
Ms. Chelimsky. I will.
Chairman Glenn. Our next witness will be Deputy Secretary of
Defense Bill Perry. . .
Senator Dorgan. Mr. Chairman, while Mr. Perry is coming for-
ward, I would like to ask that my opening statement be made a
part of the permanent record.
Chairman Glenn. It will be included in the record in its entirety.
Prepared Statement of Senator Dorgan
Mr. Chairman: I commend you for calling a hearing to re-evaluate the Nation's
strategic triad, the combination of land- and sea-based missiles and long-range
bombers that has underpinned our Nation's defense for decades.
The end of the Cold War and budget pressures together make necessary a reas-
sessment of our entire defense structure. None of us would trade the old conflict for
the new opportunities that have emerged. Also, we can all breathe a sigh of relief
that the demise of the Soviet Union and the Warsaw Pact come at a moment when
we need to get our budget deficit under control.
While the risk of nuclear confrontation has lessened, we still face unsettling devel-
opments in the former Soviet Union. Civil strife is rampant in many quarters, in-
cluding tensions in some nuclear-armed republics. Only this week, Ukrainian Prime
Minister Leonid Kuchman urged that the Ukraine became an independent nuclear
p0wer — at least temporarily. While our government is working to make the Ukraine
a nuclear-free nation, we still don't know the final outcome of those efforts.
Which means this hearing is very relevant. We need to understand what weapons
systems we still need, which ones are most cost-effective, and what is the right com-
bination of strategic forces to protect us for the years ahead.
Some have suggested that we place ever-increasing reliance on the sea-leg of the
triad. I believe this would be an expensive and risky proposal. In contrast, I believe
that we should keep a robust triad, consisting of SLBMs, up to 650 ICBMs, and
heavy bombers (including B-52s).
This arrangement provides the security that has served us well for several dec-
ades. It draws upon the flexibility of recallable bombers and retargetable ICBMs.
It keeps costs down by leveraging the investments already made in existing sys-
tems, the B-52 bombers and Minuteman III missiles. It upholds the recommenda-
tions of the Secretary of Defense in his most recent force structure to preserve stra-
tegic flexibility by retaining four northern tier missile bases.
Permit me to make a few other observations about these systems and the GAO
report.
ICBMs J .
I agree with GAO's original finding that ICBMs have been presumed to be more
vulnerable than they are. GAO noted the over-estimate of Soviet missile capability,
the likelihood that the Soviets would be deterred from a first-strike by our other
missile assets, and the failure of planners to recognize the "robust capabilities of
U.S. early warning systems to detect a Soviet ICBM attack . . . ."
I am troubled, however, that GAO questions the advisability of funding a major
service life extension for the Minuteman III missile. GAO claims it would be too
costly, unnecessary, and consumptive of test missiles.
I will raise these issues at greater length in questions later on. for now, I want
to simply say that we can achieve a stronger deterrent, with less risk, lower costs,
and a cleaner environment by balancing our missile forces between ICBMs and
SLBMs.
• For example, GAO's own analysis shows that it costs a maximum of $23 billion
to get 500 alert Minuteman III missiles good for 30 years. By contrast, spending
$58 billion would extend the life of 432 Trident missiles, but not for 30 years.
That's because the Trident submarines have to be retired after their hull life
reaches 30 years. The first submarine would thus have to be retired in 2011
and one more each year after that. Then, many of these submarines will require
a missile backfit costing $12 billion, when the expected service life of the sub-
marine will be no more than 8 years.
• Similarly, Senators Sasser and Bumpers have recommended that we stop buy-
ing D-5 Trident II missiles. We could do so reducing the number of Trident mis-
siles per submarine and using the same flight test regime for Tridents as for
the MX and Minuteman III ICBMs. The savings: $4.5 billion over 5 years.
12
• Moreover, land-based missiles afford greater reliability because they have a
much higher alert rate. The rate is 99 percent for ICBMs compared no more
than 66 percent for SLBMs — because only two-thirds of our submarines are at
sea at any given time. Reliability is still worth counting in an unstable world.
Their proven reliability also means they require fewer tests. Minuteman III
missiles now meet Joint Chiefs of Staff test standards, a fact ignored by the
GAO report.
• Single-warhead ICBM missiles also offer less attractive targets than MIRVed
SLBMs, thereby enhancing deterrence and strategic stability.
• ICBMs also measure up better on nuclear safety and environmental standards.
They scored much better than SLBMs in the prestigious Drell Panel's nuclear
safety report and produce no radioactive waste like nuclear submarines.
BOMBERS
First, the GAO report properly put to rest the myth of obsolete B-52s. As GAO
noted in its advance testimony:
We found that the B-52, whose obsolescence has been widely reported and
cited as a rationale for developing both the B-1B and B-2, is still a viable air-
craft that performs a great deal better than is generally understood.
GAO went on to say that bombers add a "critically important stabilizing character
to the overall nuclear force" because they make up the only recallable leg of the
triad.
The GAO bolstered its contention by showing that (1) the airframes and other key
components of the B-52 have reached only half their life expectancies, (2) life exten-
sion programs will assure the continuing effectiveness of the B-52, (3) B-52s com-
pare favorably with B-lBs on a broad range of measures, (4) its successful service
in the Gulf War, (5) a lessened threat from air defenses in the former Soviet Union,
and (6) the flexibility to use B-52s in either a nuclear or conventional role.
That is why I think it makes good sense to continue the flying mission of the 3-
52 and associated tankers. And this mission can be accomplished at a lower cost
by basing bombers and tankers at abases with existing missions as ICBM sites.
Once a decision is made to retain a robust ICBM force, it is much economical to
locate bombers and tankers at the same bases than to keep open a base only to han-
dle a flying mission.
In conclusion, I think the GAO report has stimulated some useful debate on the
future shape of our strategic triad. I nave suggested the outlines of the kind of force
we need to meet defense and budget criteria.
I would like to include for the record a White Paper which I prepared with my
colleagues, Senator Kent Conrad and Congressman Earl Pomeroy. This sets forth
our view that we could retain up to 650 Minuteman III missiles, as well as heavy
bombers, and still save billions of dollars, by having the right mix of strategic forces.
Chairman Glenn. Mr. Secretary, we know of your busy schedule.
We have already had one previous meeting this morning on a dif-
ferent subject, so I personally know of your schedule. I particularly
appreciate your working with us to make time available for this
hearing.
As I pointed out in my brief opening remarks a moment ago be-
fore you arrived, the hearing is not to finger-point and try and pin
anybody down. It is to try and learn the process on how the deci-
sions were made with regard to the triad.
I believe you have been given a copy of the GAO report some
time ago, and have had a chance to review it. We would appreciate
your comments on the report. We want to learn from whatever mis-
takes there may have been in the past, because in the selection of
future weapons systems we want to make them as efficiently and
on as sound a basis as possible.
We appreciate your willingness to come over this morning and
comment on the triad, giving us the benefit of your long past expe-
rience in the Pentagon. We welcome you this morning.
13
TESTIMONY OF WILLIAM J. PERRY,* DEPUTY SECRETARY OF
DEFENSE, U.S. DEPARTMENT OF DEFENSE
Mr. Perry. Thank you, Mr. Chairman.
I would like to make a few highlights or comments from an open-
ing statement and then submit a statement for the record, if I may.
Chairman Glenn. Your entire statement will be included in the
record, without objection.
Mr. Perry. I wanted to start off my observing a point which your
previous witness made, is that the Department of Defense is at the
moment in the final stages of what we call a bottom-up review,
which is a comprehensive review of the strategic forces and the
strategic strategy of the United States defense forces.
We have about 3 or 4 weeks to go to complete that review, and
we will fully brief the Congress when that is done. That will in-
clude conclusions about the nuclear forces or the strategic forces of
the United States. So to a certain extent, my discussions about nu-
clear forces are a little premature today, because we have not yet
finished that study.
Nevertheless, there are some comments I can make about that
on the basis of background work we have done and the basis of the
status of the study to date. I also will make some general com-
ments on the GAO's triad report.
Let me start off by observing that in the last decade, even since
the GAO report was started, there have been very dramatic
changes in what I would call the global environment, global na-
tional security environment. To oversimplify a very complex ques-
tion, I would say that the change in threat has two different com-
ponents.
The first is that the conventional threat from the Warsaw Pact
and of the Soviet Union has disappeared. Therefore, one of the
major threats around which our defense planning was done, which
is preparing for a blitzkrieg assault in Europe, that threat has ba-
sically gone away and the capability behind it has been very great-
ly diminished.
The second point is that, while the nuclear capability in Russia
and in some of the other former Soviet republics, while that nu-
clear capability still exists, we do not believe that we have a signifi-
cant threat of a massive nuclear attack on the United States today.
That is not because of a change in capability. That is because of
a change in the political situation.
I wanted to emphasize that difference, because the nuclear capa-
bility still remains. In the case of the conventional forces, it would
take many, many years, maybe a decade or so for any nation to re-
build the kind of a conventional military threat with which we
were faced in the mid-1980's, after a political change had occurred.
In the case of the nuclear threat, the nuclear threat could reemerge
rather quickly after a political change had been made, and that
perception of the global environment dominates our thinking on
what we should be doing relative to the nuclear strategic threat.
The second point to make about this nuclear threat is that these
tens of thousands of nuclear weapons that were in the former So-
viet Union forces, while not only still existing, but they are dis-
1 The prepared statement of Mr. Perry appears on page 49.
14
bursed in four different nations, Russia, Kazakhstan, Belarus and
Ukraine, and that increases the danger as we see it of one of these
weapons falling into the hands of a terrorist or an unfriendly gov-
ernment. Therefore, we believe that the threat of a small nuclear
incident resulting from the proliferation of nuclear weapons has ac-
tually increased in the last number of years.
Now, let me talk first, though, about the very large number of
nuclear weapons in the Soviet Union and how our strategy should
deal with that problem. I think, first and foremost, we are working
to encourage and sustain the growth of friendly democratic govern-
ments throughout the former Soviet Union. Just as I have said that
we see the threat dramatically diminish, because of the change in
the political situation, then it is very important to us to see that
friendly democratic government sustains itself now in Russia
today.
Fundamentally, that is an issue for the Russian people to decide.
We are doing what we can to sustain them, including military-to-
military contracts with the Russians. I and Secretary Aspin have
both met with the Russian Minister of Defense and the Ukrainian
Minister of Defense, and we have discussed these issues and devel-
oped cooperative programs with them.
Also, we have provided assistance to the Russian Defense Min-
istry and the Ukrainian Defense Ministry with U.S. military offi-
cials, to assist them in understanding how in a democratic society
the civilian control works in a ministry of defense.
The second point is that we are working cooperatively with other
treaty parties to bring about the implementation of the START I
and the START II Treaties. This is a very complicated situation,
because Russia has tied the entry into force of START I to
Ukraine's ratification of START I and accession to the NTP, and
the Ukrainian Rada has not yet acted on either one of those issues.
And while we are hopeful that Ukraine will ratify START and ac-
cede to the NPT, that outcome is not yet clear.
At the same time, Russia has not yet ratified START II, and its
action in that regard is linked both to the successful entry into
force of START I and it is linked in some sense to the political tur-
moil that is now under way in Russia.
In the meantime, we are proceeding to some elements of the
START I Treaty by accelerating the retirement and the elimination
of ballistic missile systems and bombers which would be covered by
START I, and we are trying to encourage both the Russians and
the Ukraines to act more quickly on START I and START II in the
entry of force, and to follow our example on accelerating the re-
moval of these weapons from the inventory. To date, we have re-
moved over 3,000 warheads from missiles whose launchers START
I would eliminate.
The third point I make is, while we are sparing no effort to sup-
port democracy in the former Soviet Union, we are aware that
there are no guarantees that that will happen. There does exist a
potential for abrupt political change. Should that occur and an un-
friendly regime come to power, they would have significant strate-
gic forces at their disposal. Accordingly — and I need to underline
this point — even though we believe that we have a friendly relation
with Russia and we do not believe they pose a nuclear threat to
15
us, we believe we need we continue to maintain a strategic force
which provides deterrent resources.
Now, the reduced threat, then, to emphasize, is not a result of
reduced capability of nuclear forces in Russia. It is a result of im-
proved political relations, and that could turn around. We hope it
will not and we are working to see that it will not. But if it turns
around, then the capability is in place and, indeed, the industrial
capabilities are still in place on which they could build even larger
nuclear forces.
So not only do we need to maintain a deterrent in place, but we
need to have some capability to reconstitute our nuclear forces
above the levels which you are now driving them to in the START
I and the START II, to hedge against the possibility that such an
unfriendly regime might not only reassert the military power, but
might begin a buildup of nuclear forces.
Precisely because such scenarios are so unattractive, we are put-
ting significant effort to strengthen democratic governments, to ac-
celerate the START I reactivations, and to achieve the earliest pos-
sible ratification of START I and START II.
Mentioning briefly the emerging threat from other nations, Sec-
retary Aspin testified to the House Armed Services Committee re-
cently that more than 20 other nations have embarked on efforts
to develop weapons of mass destruction. They do not now, as we
see it, pose a threat to the United States. It is a future threat that
we are concerned about and, as a consequence, we have placed as
a very high priority political actions we can take to prevent the fur-
ther proliferation of nuclear weapons, and that indeed will be one
of the department's highest priorities. In the meantime, we do need
to maintain some deterrent threat against regional powers who, in
spite of our best efforts, may succeed in achieving nuclear weapons.
Now, with that background on the very dramatically different
threat, and I think if how differently we size the threat today from
when I was working on this problem in the Pentagon in the late
1970's, the problem is very different, particularly in the nuclear
area, but it is still one we have to pay serious attention to. Today,
as in the late 1970's, we have to pay serious attention to how we
shape our nuclear forces to meet this threat.
We still believe we require in our nuclear forces a fully credible
deterrent capability, and by that we mean now just as we meant
in the 1970's and the 1980's, one that can complicate a potential
enemy's calculation so that he would never consider initiating an
attack.
Now, we have through the years achieved this by maintaining a
triad, that is three different elements to our nuclear forces, the
ground-based, air-based and the sea-based. The purpose of those is
not because, as some people have stated, that they give us different
capabilities, although they do, the real purpose behind the triad is
to provide the redundancy and the complementary nature that
would complicate an attacker's operational plans so that he would
not believe that he could succeed in an attack.
Nevertheless, even with this broad objective, the changing of se-
curity environment has allowed us to eliminate many of the nu-
clear systems that we have currently had in our systems. Not only
have we reduced the 3,000 warheads which I described to you, but
16
we have moved from alert that portion of our bomber force which
throughout the Cold War was kept on a state of strict alert, that
is, it was capable of being launched on very short notice.
The Minuteman II force was removed from alert status in Sep-
tember 1991, and we are currently in the position of removing all
of the missiles from their launchers. We have retired all nuclear
armed B-52G's, and are in the process of retiring all Poseidon sub-
marines carrying the C-3 and the C-4 missile. In 1991 and 1992,
President Bush reduced the number of B-2 bombers to a program
objective of 20, and terminated the advanced cruise missile and the
short-range attack missile, too. These actions were primarily taken
in the prior administration, but they are actions with which we
agree.
With START II, we are looking forward toward reducing our de-
ployed strategic forces to about 3,500 warheads. We believe that is
a sufficient number of warheads in our force, but we do believe
they must be survivable, flexible and they have offsetting
vulnerabilities by the nature of the mixture of deployment.
All of these factors are taken into consideration in the bottom-
up review, and within a month or so we will be reporting to you
what we are recommending for the exact makeup of the nuclear
forces reflecting these factors.
Now, let me comment briefly on the GAO report. It is a very for-
midable, substantial undertaking. It is in my judgment done objec-
tively, done carefully, and it will be used as a very important input.
It is being used as a very important input to our own planning of
strategic forces. We will not be so much accepting the conclusions
from it, as we will be accepting the input and the analysis in it.
I will emphasize some of the differences we have with this re-
port, but I don't want to give you the wrong perspective by empha-
sizing differences. On balance, we think it is an excellent report,
objectively done, and agree with most of the conclusions in the re-
port.
For example, if I look at the findings on that first chart on the
board there, I agree completely with the last 4 of those 6 findings.
I agree more or less with the first finding, and disagree with the
second, so 5 out of 6 isn't bad, in terms of agreement. So as I em-
phasized the 1 out of 6 that we differ with, I do not want to give
you the impression that we do not find it a very valuable report,
because we do.
Now, leaving the details of the report for the moment, let me say
that my most substantial problem with the report is the fundamen-
tal basis on which the analysis is done, which is the cost-effective-
ness analysis.
In particular, I address the method of scoring systems and scor-
ing strategic decisions by computing the program costs, life-cycle
program cost and dividing that by the surviving warheads as they
reach target. This method has a certain compelling logic to it, and
it must be an element in any analysis of the problem, but I do not
think it is a sufficient basis for making the major decisions about
our strategic forces.
My problem with it is two-fold. First of all, in consideration of
the decisions we have ahead of us now, the consideration of sunked
cost is not a useful input. The cost that has already been expended
17
really doesn't enter into our calculations on what actions and deci-
sions we are going to be making for the future. The issue ahead
of us now is what future costs have to be expended, given where
we are today, given what has been expended today and what has
already occurred.
Now, I think a more fundamental issue, though, is that the pur-
pose of our strategic forces is only indirectly to deliver survivable
warheads to targets. That is a means to an end. The end of the
strategic forces is deterrence, and the credibility of that deterrence
is the primary factor on which we have to gauge the viability of
this force.
That is not a criterion which lends itself to cost-effective analy-
sis: it does not lend itself indeed strictly to objectiveness analyses.
Therefore, while the cost per arriving warhead is an important ele-
ment that goes into our decisions and judgment on strategic forces,
it is not and it cannot be the final judgment, and, in particular, it
does not provide the basis for deciding the proper mix between air-
based systems, ground-based systems and sea-based systems.
Now, one particular example of that point has to do with looking
at the situation we have today in our sea-based forces. We are mov-
ing towards a treaty which will call for 1,750 warheads in our sea-
based forces. A very simple calculation will tell you that the most
cost-effective way of reaching that 1,750 forces is to simply stop the
program where we are today, stop the submarines in the shipyards.
We don't need to complete any more submarines, and, much less,
build new submarines, stop the missile program, we don't need to
produce any more missiles — we already have missiles with more
than 1,750 warheads — and deploy approximately 9 submarines to
sea, with the existing missiles. That would be the least costly and,
in terms of warheads delivered to target, would be equally effective
as the other alternative forces being considered.
That says nothing about the possible vulnerability of having only
two submarines deployed in each ocean. That says nothing about
the value of maintaining an industrial base either for our sub-
marines or for our missiles, in the event we ever have to reconsti-
tute our forces. The D-5 missile today is the only large long-range
missile that the United States is building, and if we shut off the
production of that missile, we will be building none. A year from
now, we could probably restart the line. In 2 years, 3 years, it be-
comes very doubtful, since all of the intellectual capital which is
formed to do that would have been disbursed by then.
Therefore, in our decisions on these forces, the point I am mak-
ing to you is that we cannot decide them simply on the basis of the
cost per delivered warhead to target. That is an important element
in the decision, but it cannot be the final decision.
A final point on the GAO report, and it is a technical point of
technical difference, I believe, and a moderately important point.
As I read the GAO report, it seemed to take the view that the
stealth characteristics embodied in the B-2 and the advanced
cruise missile were not justified by the Soviet air defense system.
That I do not agree with. I agree with the point that the Soviet air
defenses and, indeed, in general, the Soviet defense program has
not grown as dramatically in the late 1980's and early 1990's as
had been projected in the early 1980's.
18
Nevertheless, the Soviet air defense system as it now stands, So-
viet fighters and the surface-to-air missile, do pose significant
threats to non-stealthy airplanes and missiles. Indeed, I believe
this was fully demonstrated in Desert Storm. Iraq had deployed
around Baghdad some of these same fighters and some of these
same surface-to-air missiles which were developed and deployed in
the then Soviet Union, although they were a generation older than
the ones now in the Russian air defense system.
Nevertheless, when we were planning our missions in Baghdad,
we sent only two systems in the teeth of that air defense system.
One of them was the F-117 stealth fighter and the other was the
cruise missile — both the Tomahawk and the conventional version of
the ALCM. The F-117 conducted about 3,000 sorties into Baghdad,
a very dense air defense system, and lost not a single aircraft.
On the other hand, even though the other aircraft we had were
not required to penetrate this densely populated air defense sys-
tem, we still lost I believe the number is 14 aircraft in much less
stressing missions to this air defense system. So I do believe that
stealth plays a very important role in the current level of the air
defense systems, not only in Russia, but proliferated around the
world today.
The final and somewhat related comment of that is that the one
component of our strategic forces, namely our air component, has
the capability of being used, has the potential of being used either
for nuclear forces or for conventional forces, and it is my judgment
that we should be swinging most of those air-based systems over
to the conventional forces. That includes the B-1B and the B-2.
Therefore, many of the questions raised in the GAO report about
the applicability of B-1B and B-2 may turn out to be moot, if we
instead direct most of those bombers to the conventional forces.
Both the early analyses done in the Defense Department and the
analysis done by the GAO were colored to a great extent by the
Cold War nuclear situation. That is long gone. We are looking at
a very different world today, and we must plan the restructuring
of our strategic forces to face this new world, and in this new world
we have, as we see it, Russia as a friend and not an enemy. We
see the potential danger of a political reversal in Russia, at which
time we are still confronted with thousands of nuclear weapons,
and so we have to be prepared for a reversal of that threat and a
reconstitution of our own force and, finally, we are confronted with
increasing dangers of small numbers of nuclear missiles proliferat-
ing at unfriendly countries around the world.
All of those factors will be taken into account, as we structure
a nuclear force for this new world, and within a month or two we
will have this new program put together and we will be happy to
discuss it and brief it with you at that time.
Thank you very much, Mr. Chairman.
Chairman Glenn. Thank you very much.
We will follow our early bird arrivals rule this morning with 6-
minute questioning periods. Several members are here this morn-
ing.
One of the things that disturbs me most about the GAO report
is where GAO stated, "There exist systematic disparities between
estimates or claims on the likely cost performance of upgrades, the
19
actual performance of current systems, or the likely offense or de-
fense threats to these systems from the former Soviet Union."
Now, these are serious charges. If we are as far off, as GAO indi-
cates, in our estimates of what our weapons systems will do or can-
not do, then we are not doing a very good job. I know a lot of these
things did not happen on your watch, and I know that you are also
now involved, as you said, in a big reevaluation of our military, a
bottom-up review. It worries me that we not make these same mis-
takes twice here, while we are refurbishing or reanalyzing our ca-
pabilities to make decisions on future weapons systems. We have
to correct some of the errors from the past.
As I said earlier, I have backed the bombers, but not for their
nuclear delivery capability. I thought that if we got into a nuclear
war and we were exchanging nuclear weapons with somebody, with
ICBM's, SLBM's and ALCM's, and these bombs are going off, then
to expect somebody to jump into an airplane and add a couple more
nuclear explosions was not very realistic.
This emphasis on the triad has always appeared to me, quite
frankly, to be more of an emphasis on letting each service have a
piece of the nuclear pie. I do not know whether we are going to con-
tinue that or not, but I would appreciate your comment on it.
Mr. Perry. First of all, let me repeat that I found the GAO re-
port valuable, and it will be an important and well-studied input
in our restructuring the force. Also, to repeat, relative to
misestimates in the past, on that chart there the GAO has listed
six general beliefs, of which they come to different findings, and the
last four of those beliefs and findings, I agree with the GAO find-
ings, and those findings will influence our view on how we restruc-
ture the forces.
For example, that the B-52 is viable for many years to come, I
believe that is correct and that will be a factor in our consideration
of how the force will be used. It would say, for example, that we
continue to use the B-52, if we choose to do it, if we see we need
that capability as an ALCM carrier, because, while I do not believe
that the B-52 is a suitable penetrator of the air defense system of
Russia, I do believe it is very capable to deliver ALCM's, which, in
turn, are capable as penetrators.
Chairman Glenn. You mentioned Baghdad. It is a conventional
capability we are talking about there, and I do not quarrel with
that. We have to have a conventional capability and I hope we
never cross that nuclear threshold. What the GAO report basically
studied, though, was the nuclear triad.
Mr. Perry. Indeed, it did. My comment about the B-1B and the
B-2 relative to the GAO report is that it is not that I disagreed
with the findings they had there. I did not find them very relevant
to our force structure planning for strategic forces, because I am
looking to the B-1B and the B-2 primarily as an application to
conventional forces, which is the same point that you made.
Chairman Glenn. GAO states that, "No DOD evaluation exam-
ined U.S. strategic forces in this comprehensive way for at least
three decades, and the GAO was concerned to find little or no prior
recent effort by DOD to evaluate comprehensively the relative ef-
fectiveness of similar weapons systems." Now, I presume you are
20
going to remedy that with your current study, your bottom-up re-
view, is that correct?
Mr. Perry. Yes and no, Senator Glenn. The bottom-up review
will be a comprehensive assessment of the strategic forces and the
force structure. It will not have the emphasis on cost-effectiveness
analysis which was done in the GAO report, for the reasons that
I have described. We will take the cost-effectiveness analysis done
in this report as an important input to what we are doing, but that
will not be the emphasis of our study.
Chairman Glenn. In your bottom-up review that you are now
making, is this being done with a completely separate team? I am
a little concerned about who is doing all the analysis. We have De-
fense Acquisition Board, the DAB, we have the Science Board, we
have the Joint Chiefs. Who is doing this study and are we using
all of these different groups that have a part of this analysis? How
are you going about the current review?
Dr. Perry. Yes, the team that is doing this is comprised of three
major elements, the Joint Chiefs are the military element of the
chain, in policy, the Under Secretary for Policy, we have the offi-
cers concerned with nuclear systems, and, of course, the Under Sec-
retary for Acquisition has a team working on this, too.
I have to say that we have important elements of that team still
not in place yet, and it is possible that we will have some parts of
the bottom-up review done before we get the contribution of the
strategic forces, because there are some key people still missing
from the team, not yet confirmed and not yet in position. So that
may be moving a little slower than some other aspects of the bot-
tom-up review.
Chairman Glenn. You heard Ms. Chelimsky's comments about
they felt they were deliberately misled by the Pentagon. I hope we
are getting the best data and that we are not being too over-protec-
tive of internal documents that would prevent a real study of what
has happened.
I have some sympathy for their problem, because through the
years at hearings, I would ask about the B-l, because I was sup-
porting the B-l on the basis of conventional weapons delivery. We
have gotten the nuclear capabilities of the B-l as a freebie to back
up the SLBM's and ICBM's. I backed the B-l and fought on the
floor to keep the B-l. I fought very hard on the floor, and every
year I would be told, yes, we are coming along fine with the con-
ventional weapons system.
Now that the B-l is operational, we are just now getting around
to adding its conventional capabilities, after I had been told for
about 5 or 6 years, yes, we are developing and we are doing all that
work. I was flat lied to. So I have some sympathy for the GAO peo-
ple, when they say they were misled, because I was misled on the
B-l.
That is more a statement than a question, but I hope that you
are doing an honest bottom-up review this time, that everybody is
getting in the act and getting all the information they need and
that they are not being misled. My time is up, but I would appre-
ciate any short comment you could make on Ms. Chelimsky's state-
ment that they were misled by people when they were trying to do
this study.
21
Dr. Perry. Let me comment very briefly. First of all, sir, as far
as the bottom-up review is concerned, it will be as honest an effort
as we know how to make it. Secondly, relative to the GAO being
misled, I will not always agree with every conclusion in GAO re-
ports, but I will support to the death their right to have access to
honest data on which they can do their study.
Chairman Glenn. Senator Cohen?
Senator Cohen. Mr. Chairman, I would like to be a little bit
stronger on this issue, rather than sympathizing with the GAO. It
is not a question of you having been misled. You were deceived,
like the others on the Committee were deceived.
Now, we have read recent news accounts about an Air Force gen-
eral having heaped some criticism or contempt, I should say, upon
the President. But I might say that reflects in some degree an atti-
tude which has been rather prevalent over the years with a sense
of contempt being heaped upon civilian leadership.
When Senator Glenn and I served on the Intelligence Committee,
we had to go through the routine, if you don't ask the right ques-
tion, you don't get the right answer. If you do ask the right ques-
tion, you might get half of the right answer. That has led to a num-
ber of problems, I might say, Secretary Perry, whenever you have
two sets of books.
In the event that the GAO is right, there is one set for GAO and
one set for DOD. If anyone ever comes before the Congress to
present data which they knowingly know to be incorrect or false,
they ought to be prosecuted. You ought to dismiss them, but we
ought to see to it that they are prosecuted. That is fundamental to
maintaining the integrity of this system. So if we ever find a docu-
ment such as in the hands of the GAO, I think we ought to go re-
quest the Attorney General to seek to prosecute those individuals
responsible for it. So I would go a bit further than you on that
issue, Mr. Chairman, other than expressing sympathy for GAO.
GAO said, well, that is going to be cleared up now with the new
administration. That assumes that under the Bush administration
or the Reagan-Bush administrations, the Department of Defense
tolerated or encouraged the falsification of data to achieve a certain
goal. I suppose you could make the same argument under the Clin-
ton administration, there might be just as much validity, the argu-
ment that data or information will be shaped or distorted to con-
form to the cutting of budgets. So I think we have to be careful
that we not assume, just because we have a new administration,
we are necessarily going to get accurate information.
I would like to make a statement, Secretary Perry, rather than
a question. You can respond to it, if you would like. I made this
in my opening remarks about the debate over whether or not budg-
et drives strategy or strategy should in fact shape the budget.
Recently, there were news accounts that the DOD is considering
a sort of win-hold- win strategy. Some academics, at least, have
characterized that as win-lose-recover strategy, but we can leave
that for another debate. I would like to point out that when Sen-
ator Nunn requested of Walter Slocum what strategy was going to
underlie this bottom-up review, he made the following statement,
and we have this in a record submitted to the Senate:
22
"U.S. forces should be sized and structured to win two nearly si-
multaneous major regional contingencies, for example, the out-
break of war in Southwest Asia, followed by aggression in North-
east Asia. We want to avoid being placed in the position where the
United States, in effect, makes a two-war scenario more likely, by
opening up a window of opportunity for potential aggressors."
Now, that was dated just in June of this year, I believe. Now we
read that you are actually considering something quite different
than a two-regional war strategy, now a win-hold-win strategy. I
would suggest to you that such a proposal achieves the very sce-
nario that Mr. Slocum suggests we not open up a window to en-
courage and aggression by declaring that we are going to have a
window of opportunity for potential aggressors. That is a personal
opinion here.
I think what is happening is that the Department of Defense is
looking at the budgetary constraints and then devising a strategy
to conform to that. The major problem right now, and everyone will
agree, is that we don't have the capabilities to match our strategy.
But that is due primarily to a lack of lift, also perhaps a lack of
precision munitions, particularly for naval aviation, and also per-
haps some deficiencies in CQU. Nonetheless, I think what we are
seeing is an effort under way to completely remove the Gen. Powell
base force concept and replace it with this win-hold-win strategy.
I would hope that the administration would not abandon the de-
claratory policy of being able to fight two major regional conflicts
simultaneously. If we have to beef up our capabilities to match the
strategy, let us do it, or at least not solidify and consolidate the
opinion in the eyes of would-be aggressors that, indeed, we are
going to have a win-hold-win strategy, which I think would only
encourage a potential aggressor.
We now have a problem in North Korea, which is becoming more
aggressive. We still have to be concerned with Iran or Iraq, another
region. We may be faced with the possibility of a two-regional con-
flict actuality. So I would hope that you would not abandon the de-
claratory policy of having a two-regional war conflict capability,
and then to take some steps to enhance our capability to match the
strategy.
I also think that the Secretary should accept some of the rec-
ommendations about the Reserves that Gen. Powell has suggested.
I think there is a great imbalance coming about in terms of focus-
ing so much on Reserves and cutting down the active forces, and
I think that we have to really take a revaluation of the proposed
cuts by Secretary Aspin in the outyears. This may not seem quite
as ludicrous in today's climate as some might think. The House, for
example, recently passed a budget supplemental of $1.2 billion to
pay for the operations in Somalia, rather than reprogramming.
Finally, I hope we do not do what I consider to be stupid things
in this bottom-up review, such as merging the Navy's AFX with the
Air Force multi-role fighter, the F-16. If we are going to seek ways
of consolidating forces or weapons systems, it is far preferable, in
my judgment, to start consolidating F-22 and the AFX, and not the
other way around.
So I hope you will take some of these recommendations in the
spirit I offer them, and that is as aggressively as I can.
23
Dr. Perry. Thank you. May I mention two comments on that.
Senator Cohen. Now you may comment on Senator Lieberman's
time. [Laughter.]
Dr. Perry. I will try to make them fast. I think win-hold-win is
an unfortunate moniker for a complex strategy or scenario, which
is one of the three scenarios being considered in the Bottom-Up Re-
view.
You might have called World War II strategy a win-hold-win, but
that was a very vast oversimplification of what, in fact, we actually
did in that war.
But more importantly, I want to comment on the process of what
the Bottom-Up Review is about. It is not that we are coming to this
is the strategy. The key, the essence, of this Bottom-Up Review is
that we are connecting resource decisions with strategies, with al-
ternative strategies, and we will be able to present both to the
President and the Congress what sort of resources are needed to
execute alternative strategies, or alternatively, if you set the re-
sources, what kind of strategy you will be able to execute with it.
And that whole process will be laid out in front of you, and the
President and the Congress will be able to come to judgments
about how the two connect with each other. The essence of the Bot-
tom-Up Review is making the connection between strategy on the
one hand and necessary resources on the other, and we will do it
with several alternative strategies.
Chairman Glenn. Senator Dorgan?
Senator Dorgan. Mr. Chairman, thank you very much.
I would like to ask your evaluation of the discussion by the GAO
of the costs of extending the Minuteman-IIIs. They were comparing
them to the life-cycle costs of the D-5, and they indicated today
that the cost increase they used for the extension of the life cycle
of Minuteman-III has increased, I believe, from $16 billion to $23
billion. Is that the number you heard?
Dr. Perry. I can check that number for you, Senator Dorgan. I
do not have it in my head.
Senator Dorgan. Well, we do not need the exact number. But
the question is: As I heard the discussion about the D-5 versus the
Minuteman-IIIs in silos, when you put a Minuteman-III in a silo
with a single warhead, its life cycle lasts for a good long while, but
a D-5 put in a boat someplace, a submarine, lasts as long as you
have a delivery vehicle, as long as the boat lasts, and then you
have got to extend the recommended life of the boat, and I'm most
curious how you evaluate the relative costs of the sea-based ballis-
tic missiles, the D-5 and the Tridents, versus the Minuteman sin-
gle-warhead costs?
Dr. Perry. On a cost-per-warhead basis, the Minuteman with a
single warhead will be an expensive way of maintaining a system.
Even though we are extending a system that has already existed,
we lose a lot of the cost advantages of ICBMs when we go away
from the MIRVed features of that system.
The reason we are going away from the MIRVed features,
though, is because we wanted to reduce the threat from Russia,
from the Soviet Union, of MIRVed systems, and therefore we of-
fered a package of de-MIRVing our ICBMs and having them de-
MIRV theirs.
24
So I would look at the notion of bringing Minuteman-III down to
a single-warhead system as done for that reason, and the price you
pay for it is a higher cost per warhead. It makes it much less com-
petitive than a Trident
Senator Dorgan. I understand that.
Mr. Perry [continuing]. On that basis, because the Trident is a
MIRVed system.
Senator DORGAN. But would that be including the cost of extend-
ing the life cycle of the submarines?
I guess I will ask the GAO witness in some detail about this, but
I am kind of curious about the cost comparisons that were used
today. It seems to me — and I understand why we are going to a
single warhead — but having a Minuteman missile put in a silo and
lett there for some long while and with the understanding that if
you have single-warhead missiles, the risk of targeting a missile
out there with one warhead of yours to one warhead in a silo does
not make much sense, so the survivability is pretty substantial.
Dr. Perry. The survivability is good for that reason; that is right.
Senator Dorgan. And it seems to me that the cost estimates
given today by the GAO, comparing the Minutemans in place over
the 30-year life, for example, versus the D-5 in a Trident whose
life cycle we are going to have to extend, caused me some confu-
sion.
So I guess maybe I will ask those questions of the GAO.
Dr. Perry. I would suggest you refer those questions to the GAO.
My main point is that I do not think the primary basis for deciding
to keep the Minuteman force is a cost-per- warhead deliver basis.
Had we wanted to make that criterion, we probably would have
kept the three warheads in the Minuteman in the first place. It
would have been much more attractive from that point of view.
Senator DORGAN. I understand that. But the Minuteman system
over the years has been a fairly cost-effective system.
Dr. Perry. It is very cost-effective; that is right.
Senator Dorgan. And given the circumstances of what we fore-
see in the future with single warheads, it is probably cost-effective
and relatively survivable.
Dr. Perry. It is cost-effective for two reasons. First of all, it has
minimal maintenance compared with either the airborne systems
or the submarine systems. And secondly, it had this MIRVed capa-
bility which contributed greatly to the cost per warhead effective-
ness.
We have lost one of those two features, but the other one still
remains. It is still a relatively low maintenance, relatively low up-
keep system.
Senator DORGAN. Then in some of the other discussion I heard —
again, I will ask the GAO witness, but I would like your comments
on it— the discussion about the B-52s and the B-ls that talk about
its cost as a delivery system for nuclear weapons, both are also us-
able for conventional warfare. In fact, the B-l has now been des-
ignated as a conventional weapon.
I want to know whether that was considered in the evaluation,
that dual capability. But your assessment of the B-52 and the B-
1?
25
Dr. Perry. I consider the B-1B and the B-2 as primarily contrib-
utors to our conventional forces, and I think that is the way
Senator Dorgan. How about the B-52?
Dr. Perry. And the B-52H, which can carry the ALCMs, to the
extent we maintain ALCMs in our force, the B-52H will have a
unique role, and that will be strategic nuclear forces. We still have
the potential of using B-52s for conventional forces.
I think with the dedication of the B-1B and the B-2, we will not
see the merit to using the B-52 for conventional forces as well.
Senator DORGAN. But historically, we have called upon it for con-
ventional purposes, and the only point I make is, when you make
an evaluation of the relative costs of the triad, if you have a deliv-
ery system here, a B-52 that has dual-mission capabilities
Dr. Perry. That is right.
Senator Dorgan [continuing]. It seems to me that that has to be
considered
Dr. Perry. All three of those aircraft have dual capability.
Senator Dorgan. Right.
Dr. Perry. Now to make the B-1B and the B-2 — in fact, any one
of them — fully effective in the conventional capability, they need to
be outfitted with armament to make them effective in that job. So
that is a bill to be paid yet in order to get that full capability.
As the B-52, as it was used in Desert Storm, was dropping grav-
ity bombs, and it was relatively not ineffective then for that reason,
and in the Air Force's review of how they would use the B-1B or
the B-2 in a conventional war, they are imagining they would be
used with precision delivered weapons, making them much more
effective.
But that is a cost to get them outfitted for that purpose. That
is a cost not yet accommodated.
Senator Dorgan. Thank you very much.
Chairman Glenn. Senator Levin? I know Secretary Perry has to
leave right at 11:00.
I am sorry. Senator Lieberman was next.
Senator Lieberman. I took his seat. I apologize for that. If I had
known you were coming, I would not have sat here.
Senator Levin. I cannot think of anyone I would rather occupy
that seat, except for the other colleagues who are here.
Senator Lieberman. I am honored, I think. [Laughter.]
Secretary Perry, I appreciate your characteristically thoughtful
presentation. I want to pursue with you just a bit, so I am sure
that I understand what some of your constructive criticisms of the
GAO findings are.
It seems to me that what you are saying in your testimony is
that insofar as the GAO, in its conclusion that the sea leg of the
triad was the most effective, based that conclusion on cost-effective-
ness, it was missing what might be called some subjective factors
that have to do with the interrelationship of the three legs of the
triad. In a sense, this is a classic case where the whole is greater
than the sum of the parts because of both the subjective and fac-
tual impact on a potential opponent.
But in its estimate of survivability of the three legs, did the GAO
go beyond just a balance-sheet, cost-effectiveness analysis and get
to what might be called a more strategic element here, saying, if
26
I can take it further than probably is appropriate, that if you had
to choose where you, would reduce levels of threat in this new post-
Cold War environment, that it makes more sense to reduce them
in the air and land legs, as opposed to the sea, because that leg
not only has a lower cost per warhead, but also has greater surviv-
ability? . .
Dr. Perry. I think that is a fair statement of both my position
and their position, and I think I would amplify my position by say-
ing that at whatever level of warheads you have, even as you bring
them down, if I were structuring the force, I would structure them
to have several components in the force for the reasons of the di-
versity of the basing, which makes it difficult for them to be at-
tacked.
Now taking into account the conclusions of the GAO report on
the relative cost-effectiveness of the submarine versus the other
two legs, I would then say I would tend to emphasize that.
Senator LlEBERMAN. You would agree, as you have indicated
here? .
Dr. Perry. Yes, I do. I would emphasize that. And indeed, in the
force structure that has already been put together and that we are
converging to, the Trident turns out to be the dominant leg of the
triad, and it is the one on which we depend the most for the rea-
sons that were stated in the GAO report.
It is very high survivability. Even if you argue with some details
of the cost-effectiveness analysis, most of the costs on the Trident
are sunk costs at this point.
Senator LlEBERMAN. Right.
Dr. Perry. And as I said before, I am looking to the costs to be
expended yet. And on that basis, it is a very attractive alternative.
We could put the entire strategic force together with just Tri-
dents, and I recommend against that, not because the other sys-
tems are more cost-effective, but because I think the combination
of the three gives us the highest degree of confidence that we have
a system that could not be— that no one could believe they could
attack.
I do not plan to get into detail in this discussion, but to the ex-
tent you wanted to, I could describe the problems of an attacker;
on attacking submarines, it is profound. But if they were to ever
get some kind of an ASW breakthrough
Senator LlEBERMAN. Right.
Mr. Perry [continuing]. Which I do not forecast, but if they were,
then the submarine becomes the most vulnerable part of the sys-
tem.
Senator LlEBERMAN. Yes.
Dr. Perry. And the way to hedge that— and it is just a hedge-
is to have a mixture which makes it very difficult to attack.
Senator Lieberman. That is helpful. So that you are not, in
being constructively critical of the GAO problems, arguing
Dr. Perry. No.
Senator LlEBERMAN [continuing]. You are not arguing for the
triad to go on exactly as it exists now?
Dr. Perry. By no means.
Senator LlEBERMAN. And you are also not disagreeing with the
conclusion that the sea-based leg is the most survivable.
27
Dr. Perry. I believe it is the most survivable, and I believe it will
be for the foreseeable future.
Senator LiEBERMAN. But what you are cautioning us against —
and I am extrapolating here, so please correct me if needed — is
that as we go forward in post-Cold War era and adjust the triad,
we ought not to adjust it so much that we lose the land and air-
based legs. Is that fair to say?
Dr. Perry. That is fair.
Senator LiEBERMAN. Even though you would agree that perhaps
we should, in this readjustment, give somewhat more emphasis to
the sea-based leg.
Dr. Perry. I would be prepared to give quite a bit more emphasis
to the sea-based, but I would not eliminate the other two.
It would be a different question if you had asked me, would I,
from scratch, start to build the other two.
Senator LiEBERMAN. Right.
Dr. Perry. Given that we already have them, I would not elimi-
nate them. I would maintain them. On a sunk cost basis, it is still
a relatively economical thing to do, and we maintain the benefits
of the diversity.
Now unless a serious major threat reemerges, a reconstitution of
the nuclear capability, a nuclear threat from the former Soviet
Union, say, unless something like that happened, much of this dis-
cussion is academic.
Senator LiEBERMAN. Right.
Dr. Perry. We could get by very nicely with much smaller forces
and a single leg. So we are trying to hedge against the possibility
of that reemerging.
Senator LiEBERMAN. Thank you. Thank you, Mr. Chairman.
Chairman Glenn. Senator Levin, we have to vote.
Senator Levin. Shall I, then, just adjourn when I am done?
Chairman Glenn. No, go ahead. Because Secretary Perry has to
leave, let us just go ahead, and I think we have time for your ques-
tions.
Senator Levin. When I am concluded with mine, shall I just re-
cess until you return?
Chairman Glenn. Yes, we can do that.
Senator Levin. All right.
Chairman Glenn. Go ahead.
Senator Levin. Welcome, Secretary Perry.
Chairman Glenn. Just give me 30 seconds. I am very concerned
about the charge of overestimates, overestimating defenses, and
cost estimates so far off that they are not even usable. I hope you
are getting better figures out of the Services on this Bottom-Up Re-
view than what we have had in the past and what the GAO got.
Dr. Perry. One of the very positive things coming out of the Bot-
tom-Up Review is the very close working together of the military
with the new civilian team, and I am very impressed with that.
Senator Levin [presiding]. First on the question of false informa-
tion provided to the GAO, like our chairman and Senator Cohen
and, I think, every member of the Congress, we feel very strongly
about this issue. This occurred before your current position or be-
fore this Administration, so it is an awkward issue in a sense, be-
28
cause you were not here when these two events reported to us
today occurred. , ■■«-*»
One has to do with the false data. That is the word used, false ,
"purposely false data" provided, I think, on the radar cross-section
issue of the B-1B.
And then another issue which was raised by the GAO has to do
with the fact that the DOD refused to provide the GAO with reli-
ability data for the MX warhead.
So there are two kinds of issues here. Can you take a look at
both of those statements on the part of the GAO?
Dr. Perry. Yes.
Senator Levin. Will you look at those?
Dr. Perry. Yes.
Senator Levin. And would you report to this committee whether
or not, number one, to the best of your ability, they are accurate,
and number two, if so, whether the persons responsible for those
decisions are still in places of responsibility?
I am not asking you to act directly at this point, but I am asking
you to report to the committee on the factual accuracy of those two
statements of the GAO and if they are accurate, whether or not the
persons involved, are still in places of responsibility. Could you do
that?
Dr. Perry. Yes, I will do that.
[Insert for the Record]
MX Warhead Reliability Data
The facts concerning the DOD's redacting of certain portions of the report entitled
"Evaluation of Weapon System Capabilities, Peacekeeper in Minuteman Silos, are
a a follOWS*
On August 3, 1990, the Comptroller General formally requested that Secretary of
Defense Cheney provide access to the complete and unedited Top Secret/Restricted
Data report entitled "Evaluation of Weapon System Capabilities, Peacekeeper in
Minuteman Silos," dated 18 September 1989. The Comptroller General letter stated
that CINCSAC and the Department of Energy had indicated that they had no objec-
tion to release of the document to the GAO. General John T. Chain, Jr., Commander
in Chief, Strategic Air Command reviewed the Comptroller General's letter dated
August 3, 1990, and objected to the release of the report to the GAO on the basis
that "Information contained in this report is extremely sensitive, limited access,
TOP SECRET/Restricted Data, and must be carefully controlled. Release of this in-
formation would pose a grave threat to national security."
On August 31, 1990, Terrence O'Donnell, General Counsel of the Department ot
Defense, informed the Honorable Charles Bowsher, Comptroller General, that Gen-
eral Chain did object to release of the report and the Department of Energy, which
could find no evidence of having made any representation concerning release of the
report, deferred to the views of the Department of Defense on this question. Mr.
O'Donnell further reiterated his conversation with the Comptroller General, in De-
cember 1989, in which he informed Mr. Bowsher that the Secretary of Defense had
personally briefed key Congressional leaders on the matter.
Finally, on October 26, 1990, the Comptroller General reported the denial to Rich-
ard G. Darman, Director of the Office of Management and Budget. On November
28 1990, Mr. Darman made the certification pursuant to 31 U.S.C. Section 716 (d)
to the Comptroller General that, "the redacted portions of the document referred to
in your report dated October 26, 1990, which was received by the White House on
November 9, 1990, could be withheld under 5 U.S.C. Section 552 (b) (5), and that
disclosure reasonably could be expected to impair substantially the operations of the
Government." Mr. Darman's letter further described the redacted information as
"deliberative material used to make weapons allocation and targeting determina-
tions by DOD authorities charged with the oversight of the SIOP. The redacted ma-
terial consists of uniquely sensitive estimates to assist in the management of the
Nation's nuclear deterrent aimed at holding Soviet targets at risk. Reporting to Con-
gress on this extremely sensitive information appropriately has been limited to a
29
briefing by the Secretary of Defense of the leadership of the House and Senate
Armed Services Committees."
I have reviewed the facts and believe the position taken by the Department, be-
cause of the extreme sensitivity to national security, was correct.
Senator Levin. On the B-2 certification issue, I think that at
your confirmation hearing and I know at Secretary Aspin's con-
firmation hearing, I asked the question about whether or not the
Department intends to comply with the law on the issue of B-2 cer-
tification. This has to do with whether or not there will be, before
the certification is made that triggers the obligation of certain
funds, whether or not there will be actual flight tests of the inte-
grated offensive and defensive avionic systems.
Have you had a chance to look at that issue, and can you give
us your assurance that before the certification provided for by law
would be made, that, in fact, those actual flight tests of the inte-
grated offensive and defensive avionic systems will take place?
' Dr. Perry. I will personally review the certification before it is
made, and I will certainly check those points.
[Insert for the Record]
B-2 Flight Tests
The DOD will comply with section 131 of the National Defense Authorization Act
for Fiscal Year 1992 and 1993. Section 131 requires the Secretary of Defense to cer-
tify that the program for the B-2 aircraft has demonstrated sufficiently critical per-
formance characteristics from flight testing to provide a high degree of confidence
in mission accomplishment. Among these performance characteristics are offensive
and defensive avionics." The statute does not specify a requirement of flight tests
of integrated offensive and defense avionics. Whether the Secretary can make the
statutorily prescribed certification must be based on the Secretary's judgment, tak-
ing account of the flight tests conducted and, of course, the issues raised by the
GAO. It should be noted that both the B-2 offensive and defensive systems have
been through extensive laboratory and ground testing, and are now being evaluated
through flight testing. Full integration testing of both systems is scheduled to start
late in 1993, after a required software package is added, and will be completed in
late 1994. This test schedule was provided to Congress in the System Maturity Ma-
trix. To date, no integration problems have been found during any testing.
Senator Levin. Do you agree that that is a correct statement? Is
your understanding of the law the same as mine? Since I helped
write it, I do not want to assume that everybody agrees with it.
Dr. Perry. I am very careful when a lawyer asks me whether I
agree on a statement of law that he has written.
Senator Levin. Well, you are generally pretty careful anyway.
But are you familiar with it? If you are not
Dr. Perry. I have not read the law, Senator Levin, but I will, be-
lieve me.
Senator Levin. All right. If you would
Dr. Perry. Before I sign the certification, I will read the law.
Senator Levin. Well, in addition to that, would you provide for
the record your best assessment as to when you believe the Air
Force will be in a position to either make or not be able to make
that certification? What is the new timeline?
The GAO says that this is a significant period of time ahead of
us, that it cannot be made in the near future, as I remember their
written testimony or what they have told us. So can you give us
a timetable as to when you believe such certification could be made
at the earliest?
fiQ_s^Q n - Q4
30
Dr. Perry. Yes.
[Insert for the Record]
B-2 Certification
The DOD B-2 certification is planned for August 1993. The Air Force has pro-
vided B-2 testing data and analysis to OSD. The OSD staff is reviewing the infor-
mation and will provide its finding and recommendations to SECDEF in July.
Senator Levin. Now we also had a question about the warranty
provision on the B-2, and that was a provision which I also wrote
and which I believe was clearly violated by the Air Force. Are you
familiar with that issue?
Dr. Perry. Only in general terms, Senator Levin.
Senator Levin. Can you review that and tell us for the record
whether or not you believe that that provision has been complied
with, and if not, whether or not you will require compliance?
Dr. Perry. Yes. I have heard different opinions on that, and I
would have to look at that much more carefully.
Senator Levin. If you would provide that for the record as well.
We had a commitment on both of these, by the way, by the Sec-
retary at the time of his confirmation.
Dr. Perry. Yes, I am aware.
Senator Levin. I am reminding you of that.
The B-52Hs are a bomber that you have made reference to, and
I would like to ask you about B-52Gs. Is it true that they also have
some useful life left in them?
Dr. Perry. Yes, it is.
Senator Levin. Is the plan to retire them while they still have
useful life?
Dr. Perry. That will be a specific recommendation out of the
Bottom-Up Review. We have not come to that one yet, but if the
decision is to retire them, it will not be because we believe they do
not have useful airframe life.
Senator Levin. So the decision has not yet been made to retire
those B-52Gs?
Dr. Perry. Whatever has been— judgments that have been made
to date, it is on the table during the Bottom-Up Review.
Senator Levin. And so the final decision on that is still up in the
air?
Dr. Perry. That is correct.
Senator Levin. To coin a phrase.
Dr. Perry. The decision will not be, though, on the life of the air-
plane; it will be on the cost of maintaining it and the cost of outfit-
ting it to do whatever mission is prescribed for it and the value of
that mission.
Senator Levin. In any event, whatever their useful life, whatever
the decision finally is from the Bottom-Up Review, they still have
useful life?
Dr. Perry. Many years of useful life as an airframe.
Senator Levin. The Air Force was planning on reducing the
number of ICBM bases from four to three. Is that still the plan,
do you know?
Dr. Perry. I do not know.
Senator Levin. All right. Would you let us know that for the
record?
31
Dr. Perry. Yes.
[Insert for the Record]
Number of ICBM Bases
ICBMs provide a combination of accuracy, yield, range, reliability, and rapid
retargeting for maximum flexibility for the lowest operating costs. Under current
plans, the ICBM force will be reduced to 500 Minuteman III missiles deployed at
three bases. Programs are in place to maintain system reliability through 2010.
These include replacing some electronic components of Minuteman guidance sub-
systems and refurbishing solid rocket motors. Efforts to ensure that by the year
2000 all Minuteman III missiles have been downloaded to a single warhead per
START treaties continue on schedule.
Senator Levin. You have made reference to the fact that political
leadership could change in Russia, and therefore we have got to be
prepared. And I agree with you on that.
Is it also, therefore, because of that same reason imperative that
we quickly move to reduce the number of missiles in Russian
hands, and the more quickly the better because of that same rea-
son, that political leadership could revert to a prior type of leader-
ship?
Dr. Perry. Yes, that is a top priority, and to the extent we can
influence it in that direction, we would do it, including making
whatever concessions we need to make on our own forces.
Senator Levin. Have you been asked whether or not we will have
the results of the Bottom-Up Review in time for our authorization
and appropriation process? Was that asked of you this morning?
Dr. Perry. We expect to have the primary results of the Bottom-
Up Review available in early July.
Senator Levin. If you could not only provide answers to those
questions, particularly on the B-2, for the record, but provide those
within the next couple weeks, it would be most appreciated.
Dr. Perry. Yes, we can do that.
Senator Levin. Thank you. We will stand in recess until either
Senator Glenn or someone else comes back. Thank you.
[Recess.]
Chairman Glenn [presiding]. The hearing will be in order. I am
sorry for the truncated nature of the hearings, but we cannot avoid
these votes. I am sorry for the delay.
Would you expand, Ms. Chelimsky, on your comments that you
felt you were deliberately misled? That concerns me very, very
much. The DOD is now into another study and they are relying on
figures from the Services. If they are misled again and if the Serv-
ices are trying to posture for whatever particular budgetary advan-
tage there might be on certain weapons systems, now is the time
to know about things like that, so that the DOD or the other people
making these decisions are not misled again, including people here
in Congress.
Expand on that a little, if you would.
Ms. Chelimsky. Exactly, Senator. Well, there really isn't much
more to say, except that we did feel that that one passed all normal
bounds in the sense that we have a memo that says these are the
correct data, and these are the data, quite different data, that will
be sent to GAO and that have been sent to GAO.
Chairman Glenn. Is it just like that, the data are different?
Ms. Chelimsky. Yes. Oh, yes. They are quite different data.
32
Chairman Glenn. And it says that these are for internal use? In
other words, they cooked the books and
Ms. Chelimsky. In that particular case, it is the only conclusion
that I can come to, sir.
Chairman Glenn. And are the figures substantially different,
that if you use one set of figures, you come to one conclusion
Ms. Chelimsky. Yes.
Chairman Glenn [continuing]. And another set of figures, you
would come to another?
Ms. Chelimsky. They are quite different. And the real issue, the
way the thing finally worked out, was that, as you saw in the
statement, we discussed the business of the radar cross-section,
what the actual numbers were, in fact, and how only the heads-up
number had been given, so that there are several issues there.
Chairman GLENN. Is this memo classified?
Ms. Chelimsky. Yes, it is.
Chairman Glenn. We will want to look at that. I want to look
at it separately, because we may want to follow up on that.
Was this one person, or would this have been reviewed by sev-
eral people?
Ms. Chelimsky. I don't remember it, but I will give you the
memo. It is a classified memo, of course.
Chairman Glenn. Yes, I want to see it.
Ms. Chelimsky. But what we did about it, Senator, which was
the point I wanted to make to you, was that, at first they opposed
our finding on the B-1B RCS, giving us these wrong numbers.
You know how the GAO works. When we get our results, we al-
ways send them over to the Defense Department, and they com-
ment. Here they commented to us saying that our RCS finding was
not right.
And then we told them: No, we know that this is the case. And
now we have a different comment from DOD saying that, yes, we
agree with GAO's RCS finding.
So in other words, that did not represent a problem. They did not
continue with that. Once we told them we knew, that was the end
of the problem.
Chairman Glenn. I will not get into all of the details of it, since
it is classified? Is this something that was open to a considerable
interpretation where there could have been two different views of
it?
Ms. Chelimsky. I do not think so, not in my view. An RCS meas-
urement is an RCS measurement.
Chairman Glenn. Let me turn my hat around and argue on
DOD's side on this now.
Ms. Chelimsky. Yes.
Chairman Glenn. They have to take into account a worst-case
scenario. If you are in combat, you want to be prepared for the
worst thing that might happen.
Ms. Chelimsky. Yes.
Chairman Glenn. How much of their disagreement with you is
based on worst-case scenarios, as opposed to what, with 20-20
hindsight, obviously you would make some different decisions on
weapons systems, once we found out in a war that there was not
the type weapons system or defense out there? Then obviously
33
whatever we had prepared is not necessary, and that is 20-20
hindsight?
How much of it is DOD really taking the worst-case scenario and
trying to build to that?
Ms. Chelimsky. Well, it is a question of philosophy, it seems to
me. I would agree that it is always possible for somebody to be ap-
propriately conservative. Somebody can legitimately say: Well,
maybe I do not know everything about the enemy's defense; maybe
it is there, and I am just not aware of it; the data don't show every-
thing.
If that were the only issue, I would say, well, they might inflate
the threat because they were being prudent. But in this particular
case when you look at the fact that the performance of the upgrades
was also so inflated, that is not a prudent act. That is not some-
thing that you do because you are worried. That is because you are
trying to get somebody to believe that you have something which
you do not yet know you have.
I am concerned, as you are, Senator, about the Bottom-Up Re-
view. When I heard Secretary Perry say a few minutes ago that he
feels that they are not going to be doing the kind of cost-effective-
ness study that he thinks is not all you need— and I would agree
with him, that is not all you need— but that they are not going to
be doing it is of some concern to me.
I do not see why they are not producing those data and looking
at performance in the way that you would expect.
One other point I wanted to make about what he said was that
he had the idea that we had just done — and I mentioned this to
him on his way out — that we had just looked at the cost of surviv-
ing warheads generally, but we looked at cost-to-go, the life-cycle
cost-to-go, which excludes sunk costs. We took those out, because
obviously we agree with him, that really the big issue here is what
is left to spend. We cannot do anything about the sunk costs; they
are gone.
Chairman Glenn. What did you find on their look at^-well, look-
ing at your findings overall, the reader might feel that you have
been a little hard on the Air Force. Do you find anything worth
praising?
Ms. Chelimsky. Oh, yes.
Chairman Glenn. What did you find that was good?
Ms. Chelimsky. I am so glad you asked me that. I am concerned
that somebody might think that that was the case, and I am afraid
they think that.
We found, you know, an admirable job on the B-52, as I men-
tioned in my remarks, and we also feel that the air leg makes a
really important contribution, a stabilizing contribution to the
triad. As you have probably seen in GAO's statement, we feel
strongly about that.
But I think above all what you are seeing here is a change in
circumstances that has happened over the years. In other words,
if you looked at the triad 10 years ago, 15 years ago, you would
definitely say that the ICBMs had great accuracy and responsive-
ness. They were the dominant leg, along with the bombers mega-
tonnage and payload capability.
34
What has happened is that the Navy has really gotten an ability
to deal technologically with what was a difficulty of the medium;
in other words, launching missiles under sea, and they are doing
it now with similar accuracy and responsiveness to what the
ICBMs can do and with similar yield.
So what you have is a kind of improvement on the part of the
Navy. It is not the Air Force's fault that they have done that. But
the Navy additionally has the survivability that the subs have al-
ways had. So now that adds up to an extraordinarily capable and
invulnerable force. I was glad to hear that Secretary Perry agreed
with that.
Chairman Glenn. I am concerned about the process here, be-
cause DOD is into a whole new process at this time.
Ms. Chelimsky. Yes, sir.
Chairman Glenn. From your study, how do we get more ac-
countability? What process change do you think is necessary in
DOD to get more accuracy, more accountability? Part of this is just
plain honesty in reporting.
Ms. Chelimsky. Yes, which is very hard to get. There are bar-
riers. There are a number of barriers.
Chairman Glenn. If there is difficulty in getting honesty in re-
porting to get your figures, and you were getting your figures direct
from individual Services, then DOD is getting those same
figures
Ms. Chelimsky. Exactly.
Chairman Glenn [continuing]. To make their decisions on. How
do we correct that?
Ms. Chelimsky. Well, I think there are barriers in the sense that
the incentives that are there essentially tend to maximize Service
goals rather than performance goals, as you and other people said
here this morning.
And I think there is a lack of a truly independent review office
with strongly analytical capability that has the power to make a
review stick and real guarantees for independence. I think that is
a really important thing.
I think you can always go to GAO for a second review to take
a look what DOD has come up with, and we can go into the meth-
odological quality of the work that has been done.
Chairman Glenn. Let me ask about the barriers that you ran
into over there and the problems you have had. This Committee
has been very active in working very closely with GAO to expand
the IGs and to establish Chief Financial Officers, who are supposed
to evaluate how each of these Departments operates. Is it efficient;
is it not? Are we getting the right bottomline statement at the end
of the year to accomplish our objectives? All these things that are
done in any major business in the country.
Are these going to help in this regard-
Ms. Chelimsky. Well, they help with regard to-
Chairman Glenn [continuing]. Or do we need something else?
Ms. Chelimsky. Well, I think we need technical capability of the
sort that we are talking about in the studies that we did this morn-
ing and certainly in the Bottom-Up Review.
When we did our chemical warfare work with the IG's office re-
viewing it, they told us that they did not have the technical capa-
35
bility to review that sort of work; in other words, the kind of work
we have done for you this morning is different from the normal au-
diting function. This is not to say that that is not important; it is
very important, but there are some other things that you need as
well.
What concerns me the most, Senator, is that there is a sort of
attitude that a hypothesis, an opinion, an educated guess is as good
as a retrospective study. In other words, why not do both? It is fine
to have educated, very smart people tell you what they think is
going to happen. But why not look at the data that are there and
see what has, in fact, happened? Why do we not do that?
Chairman Glenn. Can you suggest any other system? Do you
suggest any other system here, other than calling on GAO or
Ms. Chelimsky. No. I think they need their own independent —
I do not think that an external office can do everything. It is just
impossible. We are just, what, 4800 people? I mean, we cannot do
all of these studies. Normally we should be looking over the De-
fense Department studies and evaluating them— that is our normal
role.
So it seems to me, they need their own independent review office
which should really be able to say something to the Service people.
My comment on that is that in the past it has been very hard for
them to gain the kind of power that they need, to be able to say,
no, this is not a good study; these are not the right data. It is very
hard for them to say that.
Chairman Glenn. Now there has been a proposal, of course, to
upgrade or to modernize Minuteman-III. Is that correct?
Ms. Chelimsky. Yes.
Chairman Glenn. Your view is that that probably is not nec-
essary and we can save some $23 billion? Is that correct?
Ms. Chelimsky. Well, our feeling is, it is not clearly justified in
the sense that it is not clear what that would get for us.
Chairman Glenn. I presume we would just leave the existing
Minuteman III missiles in place?
Ms. Chelimsky. Well, that certainly is an option that the Con-
gress could think about.
Chairman Glenn. Without going through all the expensive mod-
ernizing.
There has been some recent information about the potential Rus-
sian ASW, anti-submarine warfare capabilities. Do you still stand
by your finding, "Test and operational data show that submerged
SSBNs are much less detectable than generally understood, both
now and in future years"? Do you agree with that?
Ms. Chelimsky. We really do, Senator. I am glad you brought
that up, because Senator Cohen mentioned it earlier a little bit in
passing.
We did get a briefing on it. We know what the situation is there.
We cannot, of course, discuss it, because it is classified, but the fact
is, we have made no change in our assessment.
What I would like to explain to you is why. The basis is that
even assuming a technological breakthrough in detection, the ques-
tion is: What would it then take to make it operationally effective
against a deployed force of U.S. SSBNs? That is really the issue.
36
And we calculated the ocean areas that these sensors must
search; if the sensors are space-based, the number of satellites re-
quired for continuous coverage; the computer capabilities required
to process the data from the space-based sensors; the timeliness
and the ability of the processors to detect, identify (separating, of
course, real from false targets) locate, and track one or more
SSBNs; the need to limit the barrage area; the requirements for a
water-penetrating warhead; and the ability of the SSBN to utilize
the ocean to counter the threats.
And we conclude with Secretary Perry that this is not a likely
event.
Chairman Glenn. Were your findings on the accuracy of dif-
ferent missiles a result of test data that you had?
Ms. Chelimsky. Yes.
Chairman Glenn. It was. It was not just predictions; they were
test data?
Ms. Chelimsky. They were test data. We used test data when-
ever they were available.
Chairman Glenn. In what ways did GAO estimates on ICBM
silo vulnerability differ with the DOD estimates?
Ms. Chelimsky. Would you like to deal with that, Jonathan?
Chairman Glenn. Were there major differences there?
Dr. Tumin. The differences were — in the end, they actually con-
curred with our analysis that essentially there was not a window
of vulnerability — that, as we stated, the only way you could con-
clude that the entire ICBM silo force is extremely vulnerable is by
using only the worst-case estimates for the power of Soviet ICBMs;
that is, only by putting in the highest assumed or estimated system
reliability, the highest accuracy, and so forth, and highest yield
could you then come to the conclusion that our entire ICBM force
was vulnerable to a massive surprise attack.
In addition, however, you must take into account several of what
we considered unwarranted assumptions. For example, assuming
that the Soviets would launch a massive surprise attack without
taking into consideration both our bomber force and our submarine
force, that there would be no retaliation from those forces. So a
number of assumptions had to be used to reach the conclusion of
great vulnerability.
In their comment on this analysis in response to our final report,
DOD concurred with our analysis that ICBM vulnerability had
been overestimated.
Chairman Glenn. We are on another vote, and so we are going
to have to end here very shortly.
Do you have any additional comments from all that you have
heard this morning? I want to give the rest of you a chance to
make any comments here.
[No response.]
Chairman Glenn. Nothing? No one else? Thank you.
We are going to follow up with questions for the record. In par-
ticular I want to follow up on where you were given false informa-
tion.
Ms. Chelimsky. Yes, sir.
37
Chairman Glenn. Where we had two different sets of books, in
effect. I think that is very important, so that we do not get caught
this way again.
Ms. Chelimsky. Not again. . .
Chairman Glenn. They are making some very basic decisions
with their Bottom-Up Review right now, so I think we need to do
that right away. We will have staff getting together with you right
away.
Ms. Chelimsky. That is fine.
Chairman Glenn. You have put in a great deal of time on this,
and it is an excellent study.
Ms. Chelimsky. Thank you.
Chairman GLENN. Obviously it is something that we should be
doing and they should be doing over there
Ms. Chelimsky. Yes.
Chairman Glenn [continuing]. Going into this before, not coming
out of it afterwards.
Ms. Chelimsky. Yes.
Chairman Glenn. That is the point.
Ms. Chelimsky. Thank you.
Chairman Glenn. I think it is an excellent job, and I thank you
very much. The hearing will stand in recess.
Ms. Chelimsky. Thank you very much.
[Whereupon, at 11:34 a.m., the Committee was recessed, to re-
convene subject to the call of the chair.]
APPENDIX
Prepared Statement of Eleanor Chelimsky
Mr Chairman, I am pleased to be here today to present some of the results of
GAO's evaluation of the U.S. strategic triad. My statement is based on a set of eight
classified reports, issued in September 1992, that assessed the cost-effectiveness of
the principal weapon system upgrades in the triad's air, land, and sea legs. First
I would like to review the most important findings and conclusions of our study, and
then briefly summarize our recommendations.
BACKGROUND
In April 1990, the House Committee on Foreign Affairs asked GAO to assess the
major strategic modernization programs of the Carter, Reagan, and Bush adminis-
trations. In particular, the Committee wanted a comprehensive evaluation of the
strengths and weaknesses of these programs, and a determination of which pro-
posed upgrades appeared to be the most cost-effective. This required us to make
analytic comparisons between deployed and proposed weapon systems, and across
strategic programs in all three legs of the triad, taking into account the threat they
were intended to address and the arms control agreements that would likely con-
strain or curtail them. In conducting our literature search for prior similar studies,
we found that no Department of Defense (DOD) evaluation had examined U.S. stra-
tegic forces in this comprehensive way for at least 3 decades.
The nuclear weapon systems and proposed upgrades included in our evaluation
were, for the air leg: the B-52G and B-52H, B-1B and B-2 bombers; the ALCM,
ACM, SRAM A, and SRAM II missiles; for the land leg: Minuteman II and Minute-
man III ICBMs; the Peacekeeper, Peacekeeper Rail Garrison, and the small ICBM;
and for the sea leg: the C-4 and D-5 SLBMs on Lafayette and Ohio-class nuclear-
powered ballistic missile submarines (SSBNs). We assessed all systems under a full
range of threat scenarios, moving from total surprise attack to strategic warning.
To compare system costs across strategic program upgrades, our unit of analysis
was the 30-year life-cycle (that is, we included not just R&D and procurement, but
also operations and support costs for every system). To compare system effectiveness,
we used seven different measures: (1) survivability against both offensive and defen-
sive threats, for both platforms and weapons (for example, submarines and their
ballistic missiles; bombers and their missiles); (2) delivery system performance (that
is, accuracy, range, and payload); (3) warhead yield and reliability (that is, the prob-
ability that the warhead will detonate as intended); (4) weapon system reliability
(that is, the combined reliability of all the component processes from platform
launch to warhead detonation); (5) flexibility across a number of dimensions, includ-
ing retargeting, recall, and impact on arms control; (6) communications (for exam-
ple, connectivity between command authority and platforms); and (7) responsiveness
(that is, alert rate and time-to-target).
Establishing these comparisons required a good deal of test, performance, and cost
data. In the great majority of cases, we benefited from the able assistance of the
Defense Department. However, in one very important instance, we were denied ac-
cess to data critical to establishing the reliability of the Peacekeeper warhead.
We organized our comparisons around seven policy questions, each presented in
a separate volume of the triad series, along with a summary report. The questions
are as follows:
(1) How vulnerable are U.S. SSBNs?
(2) How vulnerable are silo-based ICBMs?
(3) What is the relative effectiveness of ICBMs versus SLBMs?
(4) What improved capabilities do the air leg's proposed upgrades provide, relative
to existing systems?
(5) What are the comparative costs of the proposed upgrades?
(39)
40
(6) What capabilities exist within the triad for addressing a threat posed by stra-
tegic relocatable targets (SRTs)?
(7) What strategic capabilities exist in France and the United Kingdom?
Our evaluative approach was thus designed to analytically compare the major
strategic weapon system delivery platforms, missiles, and warheads, incorporating
arms control, threat, cost, and performance considerations. It was also intended to
provide a comprehensive framework that would permit ongoing and future calcula-
tions of the number and structure of strategic forces likely to be the most cost-effec-
tive under differing arms control and threat configurations. We believe this latter
capability is important in view of continuing arms reduction agreements, evolving
uncertainties in the Commonwealth of Independent States (CIS), a changing set of
threats to our security, and consequent requirements for potentially different mixes
of weapon systems. In addition, the high cost of new procurement in a period of di-
minishing resources, and the recurring need to decide which weapon systems are
most desirable to retain, make it critical for policymakers to know with confidence
which weapon systems operate as intended, which actually possess the performance
characteristics commonly attributed to them, and how they compare in a variety of
measures.
It goes without saying that this short statement can present only highlights of
GAO's eight reports. Also, because much of the data and many details of the issues
we examined are classified, both this statement and my responses to your questions
are constrained by security requirements. (I would, however, welcome the oppor-
tunity to respond to questions on classified issues in executive session.)
Now let me turn to our findings.
MAJOR GAO FINDINGS
Findings on the Air Led
First, in examining the flight-test performance of the B-2, we found problems in-
volving the all-important stealth characteristics intended to reduce the bomber's de-
tectability. These problems have also been reported by the Air Force: that is, the
B-2 did not perform in 1991 tests as required. More recently, however, the Air
Force reports that testing has progressed far enough to satisfy the various certifi-
cation requirements imposed by the Congress on the B-2, in light of the 1991 test
Sroblems. Nonetheless, only one quarter of the B-2's flight testing hours had been
own as of May 1993, and past experience has been that important and costly prob-
lems have emerged, not only at late stages of flight testing, but also after deploy-
ment. Further, as we reported to the Congress in December 1992, the certification
to be made would be issued without benefit of actual B-2 flight tests of integrated
offensive and defensive avionics. Given the limited number and nature of the tests
held to date, it is difficult to have a high degree of confidence in overall B-2 per-
formance at the present time.
Indeed, the history of the less technologically ambitious B-1B reinforces the prin-
ciple that final evaluation of weapon system performance should be reserved until
all operational testing is completed. For the B-1B, we found that although DOD has
reported success in reducing its radar cross section (RCS), the measurement cited
by DOD is questionable from the viewpoints of both representativeness and accu-
racy. First, even though the B-lB's RCS has been measured from all angles, only
the head-on RCS measurement was reported. That is, the side and rear measure-
ments, both of which are very much larger, have not been publicly presented. Since
radar intercepts can occur at any angle, head-on data alone cannot be a representa-
tive measure of detectability. Second, we found that even the head-on measurement
reported did not correspond to actual test results.
In other performance areas, the B-1B has had a long history of test and oper-
ational shortcomings. It has been grounded numerous times; its electronic counter-
measures continue to be a major problem; its flight controls have needed significant
modifications; and we found its reliability and range to be areas for serious concern
that require further testing. These persistent deficiencies may not be easily or
quickly resolvable, and substantial additional costs may be involved.
Because of this B-1B history, the cost of the B-2 program is of particular concern,
especially given the paucity of data on operational and support costs. Currently, the
E rejected cost to acquire the 20-aircraft program is $44.4 billion. Of this total, $2.8
illion was authorized but "fenced" by the Congress for the last five B-2s requested
by the Bush Administration. It is important to note that the five aircraft together
would be able to deliver only about 2.3 percent of the total number of strategic war-
heads permitted under the START II agreement, and, as discussed below, the B-
41
2's lifecycle cost per arriving warhead substantially exceeds that for weapon systems
in either of the other legs.1
On the other hand, we found that the B-52, whose obsolescence has been widely
reported and cited as a rationale for developing both the B-1B and B-2, is still a
viable aircraft that performs a great deal better than is generally understood. Air
Force flight hour data show that, as of 1990, the airframes and other key structural
components of both the B-52G and the B-52H had reached only about half then-
life expectancies. In addition, the Air Force has been performing numerous modifica-
tion programs over the life of the B-52 to assure its continuing effectiveness; these
include new offensive and defensive avionics, new communications equipment, new
missile launcher racks to provide cruise missile compatibility, and various programs
to enhance reliability and maintainability. Further, comparisons of data on multiple
measures of effectiveness show that the B-52 compares favorably to the newer B-
1B which has shown deficiencies on a number of important performance dimensions
(for example, reliability or electronic countermeasures). Both models of the B-52
have continuing capability— the B-52G as a cruise missile carrier and the B-52H
as a strategic penetrating bomber— and both should remain usable aircraft for many
years to come, in both conventional and strategic roles. Indeed, the entire B-1B
force was grounded for the duration of the Gulf War, while the B-52s were major
participants. . , D „
Also, in further examining the rationales supporting the need tor the 07A we
found that the Soviet air defense threat, like the B-52's obsolescence had been
overestimated.2 Evaluation of the data over the period 1972-1991 showed this clear-
ly with regard to both the number and the effectiveness of Soviet an defenses
against existing U.S. bombers and their weapons. Today, the breakup of the Soviet
Union the rivalries among the CIS states, and economic conditions within the Com-
monwealth suggest that current air defenses are more likely to degrade than im-
prove. In short, the Soviet air defense threat that the B-2 had been created to ad-
dress was never in fact deployed. .
With regard to air leg armaments, we found that the actual range ot the Air
Launched Cruise Missile (ALCM) was better than what had been reported. This
means that the improvement in range to be brought by the ALCM upgrade, that
is, the Advanced Cruise Missile (ACM), was only slightly greater than the older
ALCM's demonstrated capability. We also found that the improvement in accuracy
offered by the ACM appears to have little real operational significance.
Findings on the Land and Sea Legs
We found that the Soviet threat to the weapon systems of the land and sea legs
had also been overstated. For the sea leg, this was reflected in unsubstantiated alle-
gations about likely future breakthroughs in Soviet submarine detection tech-
nologies, along with underestimation of the performance and capabilities of our own
nuclear-powered ballistic missile submarines. The projected threat to the sea leg
was, however, used frequently as a justification for costly modernizations in the
other legs to "hedge" against SSBN vulnerability. Our specific finding, based on
operational test results, was that submerged SSBNs are even less detectable than
is generally understood, and that there appear to be no current or long-term tech-
nologies that would change this. Moreover, even if such technologies did exist, test
and operational data show that the survivability of the SSBN fleet would not be in
question. ,
In the case of the land leg, we found that the claimed "window of vulnerability
caused by improved Soviet missile capability against our silo-based ICBMs was
overstated on three counts. First, it did not recognize the existence of sea and air
leg deterrence— that is, the likelihood that the Soviets would hesitate to launch an
all-out attack on the ICBM silos, given their inability to target submerged U.S.
SSBNs or on-alert bombers and their thousands of warheads that could be expected
to retaliate. Second, the logic behind the claim assumed only the highest estimates
for such key Soviet missile performance dimensions as accuracy, yield, and reliabil-
ity, while at the same time discounting very substantial uncertainties about per-
formance that could not have been resolved short of nuclear conflict. Third, it ig-
1 Our analysis of the B-2 focused on its originally intended strategic-nuclear mission; however,
since the breakup of the Soviet Union, the Air Force has articulated a conventional role for the
aircraft. This newer rationale for the bomber is addressed in our February 1993 report, Strategic
Bombers: Adding Conventional Capabilities Will Be Complex Time-Consuming and Costly,
(GAO/NSIAD-93-45). In this report, we found that, given the very incomplete nature of the
flight test program, it is premature to confirm the actual operational capabilities of the aircraft
in a conventional role.
2 Indeed, our analysis revealed a fairly large number of areas in which the available data did
not support many conventionally-held beliefs. (See appendix I for a display of these issues.)
42
nored the capabilities of U.S. early warning systems to detect a Soviet ICBM attack
and, thereby, allow a reasonably rapid response.
With respect to ICBM performance, we found much more uncertainty on a num-
ber of dimensions than expected. Within the triad, the land leg's ICBMs have long
been perceived as having the highest reliability of any weapon system, as well as
the greatest accuracy. But we found, using test data, that accuracy estimates for
the Peacekeeper — the lead ICBM system — were based on a very limited number of
test shots, some using operationally unrepresentative software or hardware. As of
early 1992, accuracy estimates were based on data from fewer than 25 launches, of
which the first 18 combined developmental and operational elements. This alone
creates considerable uncertainty in accuracy claims derived from these results. Sec-
ond, because DOD refused to release critical data on Peacekeeper warhead reliabil-
ity, we cannot validate DOD's high estimates for it. Third, to lower costs, SAC re-
duced the Peacekeeper's test rate from 8 to 3 shots per year, which further dimin-
ishes confidence in any future estimates of the system's performance. (Similarly, the
test rate of the Minuteman HI system was also reduced — from 7 to 4 shots per
?ear — thereby also decreasing, over time, the credibility of performance estimates.)
n sum, uncertainty in the estimates for the Peacekeeper is created by a combina-
tion of inadequate evidence, insufficient test rates, and gaps in the data.
In contrast, we found that the sea leg's performance has been understated (or
poorly understood) on a number of critical dimensions. Test and operational patrol
data show that the speed and reliability of day-to-day communications to sub-
merged, deployed SSBNs' are far better than widely believed, and about the equal
in speed and reliability of communications to ICBM silos. Yet conventional wisdom
gives much higher marks to ICBM command and control responsiveness than to
that of submarines. In point of fact, SSBNs are in essentially constant communica-
tion with national command authorities and, depending on the scenario, SLBMs
from submarine platforms would be almost as prompt as ICBMs in hitting enemy
targets.
Other test data show that the accuracy and reliability of the Navy's D-5 SLBM
are about equal to DOD's best estimates for the Peacekeeper. Further, its warhead
has a higher yield than the Peacekeeper's. In short, we estimate that the D-5 has
a hard target kill capability about equal to the Peacekeeper's, while its platforms
remain virtually undetectable, unlike easily located silos.
Findings on Triad System Costs
We compared the 30-year life-cycle costs of the major triad system upgrades, tak-
ing into account the whole range of attack scenarios, and using DOD's own estimates
for the performance of each major upgrade, whether or not the current test data
supported these (high/best) estimates. Measured in terms of life-cycle costs per ar-
riving warhead, the B-2 would cost between 2V2 and 5 times more than the D-5/
Ohio system under any attack scenario, depending on the number of warheads on
the D-5. (These estimates are conservative in that they assume the B-2 will be as
effective as planned by DOD and that costs will not grow, whereas the cost, test,
and operational performance data on the D-5/Ohio system are considerably more re-
liable and complete.) When we compared the upgraded/de-MIRVed Minuteman III
system — now being proposed by the Air Force — to the life-cycle cost-to-go per arriv-
ing warhead for the D-5/Ohio system, we found they were almost identical, but with
the significant performance advantage for the latter of being based on submerged,
essentially invulnerable submarines.3
Findings on DOD's Evaluations of Its Strategic Programs
In comparing performance and cost across the legs and weapon systems of the
triad, we were concerned to find little or no prior recent effort by DOD to do what
we were doing — that is, evaluate comprehensively the relative effectiveness of simi-
lar weapon systems. Yet such agency evaluation is critical if limited budget dollars
are to be concentrated on programs that are both needed and effective.
With regard to proposed upgrades, we found many instances of dubious support
for claims of their high performance; insufficient and often unrealistic testing; un-
derstated cost; incomplete or unrepresentative reporting; lack of systematic compari-
son against the systems they were to replace; and unconvincing rationales for their
development in the first place. Where mature programs were concerned, on the
3 This comparative coat estimate is biased against the D-5/Ohios, since the Air Force has in-
formed us that the Minuteman III cost estimate on which it is based — and which they had pro-
vided to us — actually understated lifecycle costs by nearly 40 percent, for maintaining the Min-
uteman III force through the year 2020. Further, even the revised Minuteman III cost estimate
may be overly optimistic, in that it assumes that maintenance costs do not increase over the
2010-2020 decade, compared to the previous one.
43
other hand, we often found that their performance was understated and that inap-
propriate claims of obsolescence had been made.
Specifically, we found that the vulnerability of our B-52s, submarines, and silo-
based ICBMs to a Soviet threat had been overstated; that performance claimed for
the B-2 is as yet unproven; that B-1B, ACM, and Peacekeeper capabilities were
often inflated; that costs for strategic systems generally were incomplete (operating
and support costs having typically gone unreported); and that the performances of
B-52s and SSBNs were consistently understated.
We looked for assessments systematically comparing proposed upgrades against
the weapon systems they were intended to replace and found none in the cases of
the B-2, the B-1B, and ACM; we found insufficient test samples for the B-1B,
ALCM, ACM, SRAM A, and Peacekeeper; and we found many examples of unrealis-
tic testing for the B-1B, ALCM, ACM, and Peacekeeper.
Perhaps the most important point here is that comparative evaluation across the
three legs of the triad — and between individual weapon systems and their proposed
upgrades — has been signally lacking. This is unfortunate because it deprives policy-
makers in both the executive branch and the Congress of information they need for
making decisions involving hundreds of billions of dollars. (The life-cycle costs for
triad modernization stood at about $350 billion in 1990.) Examples of generic areas
in which we found significant knowledge gaps are given in appendix II.
This is not to argue that narrower evaluation should not also be done, and done
realistically and rigorously. Indeed, we have seen some examples of excellent work
of this type, including the evaluations of SLBM and SSBN performance produced
for DOD by the Johns Hopkins University Applied Physics Laboratory. But examin-
ing whether a weapon system meets its specifications cannot get at larger evalua-
tive questions like (1) whether the mission to be addressed by a proposed new sys-
tem is already adequately handled by capabilities existing elsewhere in the triad,
or (2) whether that new system has the capability to improve significantly on exist-
ing performance, and at what relative cost.
MAJOR GAO CONCLUSIONS
Based on the comparative findings presented above and the analysis conducted for
our studies, we conclude that, on balance, the evidence shows the sea leg to be the
strongest, most cost-effective component of the triad under a range of scenarios.
A second conclusion concerns the role of the air leg in the context of the triad.
Because strategic bombers are recallable (as missiles are not), and because they are
virtually incapable of effecting a surprise attack, they add a critically important sta-
bilizing character to the overall nuclear force. This is not to argue the case for any
particular bomber, but rather to draw attention to the contribution of the air leg
as such.
Finally, on the subject of evaluation, we are, of course, concerned by the multiple
individual flaws and failures we found in areas like threat forecasting, testing, and
reporting. However, we are even more concerned by the dearth of comparative stud-
ies that are needed to show whether a proposed system is justified in terms of the
threat it faces, its performance capabilities vis-a-vis other systems, and its relative
costs.
GAO RECOMMENDATIONS
Based on these findings and conclusions, we make- the following six recommenda-
tions to the Congress.
• With respect to whether five more B-2s should be procured, we find no strategic
grounds for acquiring them. Regarding the more recent justification that the
program is needed to fill a conventional role, we find that adding such capabil-
ity to the B-2 strategic bomber design will be complex, time consuming, and
extremely costly. In addition, its capability to perform either of its intended con-
ventional and strategic missions remains unproven, and other alternatives
exist.
• More operational testing of the B-1B is needed to verify that scheduled im-
provements in reliability and electronic countermeasures are achieved, and to
remove remaining uncertainties concerning range performance.
• On Minuteman III, we question the advisability of funding extensive major life-
service upgrades for this force because the cost-effectiveness of such an effort
is not obvious. There are three reasons for this: the Air Force's estimated $23
billion as the price tag for upgrading and maintaining it through the year 2020;
the fact of a reduced-threat environment, now and in the foreseeable future; and
44
the likelihood that substantive modifications would require robust flight test
programs that would quickly use up limited test assets.
• Given the importance of the D-5 missile to the sea leg of the triad, and given
the importance of flight testing to achieve an understanding of missile perform-
ance, adequate D-5 SLBM flight testing should continue. The D-5 test rate
should not be cut from levels required to confidently assess weapon system ca-
pability, as has occurred with the Peacekeeper and Minuteman forces.
• On the ACM, we concurred with the September 1991 decision to cap production
at 520 missiles, rather than funding an additional 120, given that ACM pro-
vides little operationally significant improvement over the older ALCM. We
would also concur with a decision to cap the program at an even lower level.
However, to ensure the effectiveness of the cruise missile inventory, we see a
need to hold more realistic flight tests of ALCM's survivability and of both
ALCM's and ACM's performance over terrain that has not been pretested.
• On evaluation, we would reiterate the Comptroller General's recent suggestion
that the Congress consider setting aside hearing time each year for Federal
agencies — in this case, DOD — to present the results of requested evaluations,
studies, and audits. We believe that more frequent congressional hearings on
weapons performance, combined with regular congressionally-mandated evalua-
tions, would provide more of an incentive to DOD both to emphasize the quality
and usefulness of its analyses, and to undertake the critically needed compara-
tive evaluations.
The Defense Department's response to our series of reports is to "partially concur"
on some of our findings, and to disagree with others. Where appropriate, we modi-
fied our language based on DOD's comments or on new data they supplied. In other
areas, however, we must continue to disagree. For example, one DOD concern cen-
ters around our reanalysis of their estimates of the Soviet air defense threat: they
point out that past intelligence projections will invariably show divergences from
more current ones.
While it is true that projection errors are always to be expected, they normally
occur in both directions, either overstatement or understatement. What we found,
however, is that DOD's threat projections were rarely if ever understated, but rath-
er, in the vast majority of cases, greatly overstated. Because the error was always
in one direction, this undermines the explanation of random divergence; in particu-
lar, the effect was to make the threat loom larger than the data could support.
We would agree with the Defense Department that the international atmosphere
of our time is greatly altered and that the strategic threat has changed, but we
would also note that because nuclear weapons remain in force in many places in
the world, our new defense posture must take these realities — and especially the re-
alities of performance and cost — into account. The fact that the threat has changed
does not mean that sound information is no longer needed on the triad and its com-
ponent parts. Indeed, decisions on procurements, appropriations, and budgetary
realignments will continue to require the very best possible evaluative analysis.
It may be worth noting that while DOD differed with us on some findings, the
Bush administration's actions in fact mirrored some of our major recommendations
in early drafts of our triad capping report. For example, we questioned the need for
either SICBM or Peacekeeper rail garrison; both were cancelled by President Bush.
We found no need for even four more B-2s; that force was cut from 75 to 20 by
President Bush. We also questioned the need for the ACM, and noted that insuffi-
cient tests of the Minuteman lis precluded any confidence in estimates of the mis-
sile's reliability; President Bush cut the ACM buy from 1,000 to 520, and decommis-
sioned the entire Minuteman II force.
We hope the findings we have presented here on weapon system and cross-leg
cost-effectiveness will assist both DOD and the Congress in deliberating and deter-
mining the future size and structure of a nuclear force that (1) integrates our most
effective weapon systems into a leaner, less costly whole, and (2) meets the Nation's
strategic security requirements for many years to come.
Mr. Chairman, that concludes my remarks. I would be happy to answer any ques-
tions you or the Committee may have.
45
APPENDIX I
APPENDIX I
Table 1.1: The Air Lea: Beliefs Versus Findings1
££LX££ FtNPtW?
1. On Air Base Survivability
Bombers at bases have been
vulnerable to surprise
Soviet attack.
2. On Penetration Survivability
Soviet air defenses have grown
dramatically.
Soviet SAMs and interceptors
are very effective.
B-2 is needed to preserve
the penetrating bomber role.
ACM is needed to overcome low
ALCM survivability.
3. On Target Coverage
Detectability and slowness make
the air-leg "stabilizing."
B-1B and B-2 have sufficient
range for their strategic
mission requirements.
Bombers are readily recallable
and retargetable under any
scenario, including nuclear
war.
B-2 is needed for SRT missions.
4. On Obsolescence
B-52 age mandates replacement.
The data show surprise attack
to have been extremely
unlikely.
High growth did not occur.
Combat experience and
intelligence assessments
indicate lesser capabilities.
Data show B-1B and B-52H can
continue to be survivable
penetrators.
Tests did not demonstrate
low ALCM survivability.
Available data
support this belief.
Insufficient evidence to
support this belief; reliable
test data are lacking.
Nuclear effects and jamming
are likely to degrade C3,
thus limiting recallability
and retargeting.
Analysis shows that
no special capability
exists or is foreseen.
Air Force data show B-52G & H
viability for many years to come.
10nly selected material on beliefs versus findings is presented
here; classified information has been deleted.
22
46
APPENDIX I
APPENDIX I
Table 1.2: The Land Leg
EEilEZ
1. On ICBM Base Survivability
Beliefs Versus Findings2
FINDING
Silo-based ICBMs have been
highly vulnerable to massive,
surprise Soviet attack.
2. On Penetration Survivability
ICBMs face no effective ABM
defenses.
Claims for high vulnerability
were based on worst-case
estimates of Soviet ICBM
capabilities, as well as other
questionable assumptions.
Available data support this
view.
3. On Target Coverage
ICBM C3 is prompt, reliable,
and has great redundancy.
ICBMs can launch promptly
after receipt of orders
for attack.
Available data generally
support this perception.
Available data support this
conclusion, but are based on
launches from test silos and
simulated electronic launch
tests.
DOD's refusal to provide
critical reliability data
and insufficient operational
tests reduce the level of
confidence in Peacekeeper's
performance estimates.
Rail garrison Peacekeepers Insufficient data to support
and mobile SICBMs would have this belief,
the same accuracy and reliability
as ICBMs in silos.
Peacekeeper is very accurate
and very reliable.
20nly selected material on beliefs versus findings is presented
here; classified information has been deleted.
23
47
APPENDIX I
APPENDIX I
Table 1.3: The Sea Lea: Beliefs Versus Findings3
BELIEF FINDING
1. On Survivability
While submerged SSBNs are
currently hard to detect,
a breakthrough in detection
technology that will threaten
them is possible in the
future .
2. On Penetration Survivability
SLBMs face no effective ABM
defenses.
3. On Target Coverage
C3 to SSBNs is much slower and
much less reliable than to
ICBM silos.
SLBMs cannot be used against
time urgent targets due to a
combination of slow C3
and launch procedures.
SLBMs cannot effectively attack
the hardest category of Soviet
targets due to insufficient
accuracy .
Range and deployment area
limitations may weaken
sea leg accuracy and
survivability.
No current, near- or far-term
submarine detection
technologies, potential
applications, or Soviet
capabilities would be
effective in reliably locating
a single submerged, deployed
U.S. SSBN, much less the
entire fleet.
Available data support this
assumption.
Data show C3 to SSBNs
is about as prompt and as
reliable as to ICBM silos,
under a range of conditions.
Compared to ICBMs,
no operationally meaningful
difference in time to target
was found. Arms control
agreements will severely
reduce the number of "time-
urgent" Soviet ICBM targets.
Test data show that D-5 SLBMs
do in fact have this
capability.
SSBN patrol areas and D-5
range and estimated accuracy
impose no such limitations.
3Only selected material on beliefs versus findings is presented
here; classified information has been deleted.
24
48
APPENDIX II
APPENDIX II
Table II. 1: _ GAO's Findings on Significant Knowledge Limitations
Vis-a-Vis Three Dimensions of Strategic Weapons System
Assessment
Air
Land
Sea
B-2
B-1B
B-52
Peace-
keeper
MM III
D-5/
Ohio
Threat*
X
X
X
X
X
X
Performance'
X
X
X
X
Testing6
X
X
X
X
"Threat or performance has been incorrectly reported on at least
one significant dimension.
Operational testing has experienced a significant qualitative or
quantitative problem or limitation.
25
49
Prepared Statement of William J. Perry
Good morning Mr. Chairman and Members of the Committee. Thank you for the
opportunity to present my views in conjunction with the committee's hearing on the
"Evaluation of the Viability of the U.S. Strategic Nuclear Triad." As you know, the
Department is currently reviewing the full range of DOD force structure options. We
expect to. complete that review shortly. Since it is ongoing, I will defer comment
on the specifics of our work. However, I do want to take this opportunity to discuss
the structure of our nuclear forces in the context of today's global environment I
would also like to make some general comments on the GAO's Triad report, al-
though the Department has not yet formally completed its comments on the GAO
final report and findings.
Global Environment for Strategic Nuclear Forces
Today's changing security environment confronts us with significant uncertainties
and challenges in our force planning The disappearance of the Warsaw Pact and the
subsequent breakup of the Soviet Union dramatically altered the global security
landscape in a very positive way. The conventional threat to our NATO allies has
largely disappeared and, in light of our continually improving relationship with the
Russian government, the threat of a massive nuclear attack on the United States
is at an all time low. There are nevertheless, issues with which we have to deal;
the nuclear weapons deployed by the former Soviet Union still number in the tens
of thousands, and many of these are deployed in Ukraine, Kazakhstan and Belarus;
the threat of FSU nuclear weapons falling into the hands of terrorists or unfriendly
governments still remains — a problem we have dubbed "loose nucs"; in addition, an
increasing number of countries possess, or are seeking to acquire, technologies asso-
ciated with weapons of mass destruction. I would like to describe to you our strategy
for dealing with these threats to American security.
Nuclear Weapons in the Former Soviet Union
The nuclear weapons of the former Soviet Union represent the most proximate of
the potential nuclear threats we face. We are dealing with this on four fronts.
First, and foremost, we are working to encourage and sustain the growth of
friendly democratic governments throughout the former Soviet Union. This involves
a wide range of government to government contacts and programs in which the De-
fense Department has a large and important role. As you may know, Secretary
Aspin has just returned from meetings with Russian Minister of Defense Grachev
and Ukrainian Minister of Defense Morozov. In each of those meetings the Sec-
retary expressed our strong desire to build links between the DOD and their Min-
istries of Defense; these will help expose Russian and Ukrainian officers and offi-
cials— including the younger ones who will be tomorrow's leaders — to our democratic
way of life and will help demonstrate to them how the military and defense proc-
esses function in a free society under civilian control.
Second, we are working cooperatively with the other Treaty parties to bring about
the implementation of the START I and START II treaties. As you know, Russia
has tied the entry into force of START I to Ukaine's ratification of START I and
accession to the Nuclear Non-Proliferation treaty as a non-nuclear weapons state;
the Ukrainian Rada continues to debate these treaties, and while we are hopeful
that Ukraine will ratify START and accede to the NPT as a non-nuclear weapons
state, the outcome is not clear. Russia has not yet ratified START II, and its action
in that regard is linked both to successful entry into force of START I, and to the
resolution of the political turmoil in Moscow. So there is much to be done on the
diplomatic front before these landmark treaties are fully in place. However, we want
to see the earliest start of the process of deactivating and dismantling the substan-
tial nuclear arsenal that once belonged to the former Soviet Union. We have set an
example for our other treaty partners by accelerating the retirement and elimi-
nation of the ballistic missile systems and bombers which START I requires us to
destroy. We believe our partners can benefit both themselves and the United States
by following our lead. To this end, we are working with Moscow and Kiev to develop
means to help accelerate the removal of warheads from ballistic missile systems
START would eliminate, and to assist subsequently in the removal of the missiles
from their launchers and the destruction of those launchers. It is worth noting that
every former Soviet nuclear warhead which is destroyed is one less nuclear warhead
which terrorists can seize; additionally, we are working with Russia to ensure that
the nuclear material from the dismantled warheads is accurately accounted for and
securely controlled — again keeping this material out of the hands of terrorists. In
this manner, accelerated deactivation and implementation can serve two security
goals at once.
50
As I noted a moment ago, our own record in early deactivation of strategic forces
is a good one. To date the United States has removed over three thousand warheads
from missiles whose launchers START I would eliminate— representing 80 percent
of the total number of such warheads; we will have fully completed this task by the
end of calendar year 1994. In doing so, we are reinforcing in the minds of our
START partners that we view them as friends and that they have nothing to fear
from us; by the same token, we seek to encourage them to accept our offers of assist-
ance and similarly accelerate the draw down of their strategic systems.
Third, while we are sparing no effort to support democracy in the FSU and to
achieve the reduction including on an accelerated basis — of START I associated sys-
tems, we are also aware that there are no guarantees. Despite our best efforts, there
still exists the potential for abrupt political change in the former Soviet Union.
Should this occur, and unfriendly regimes come to power, they would have signifi-
cant strategic forces at their disposal. Accordingly, we need to continue to maintain
a strategic force which provides the deterrent resources we would need in such a
situation.
Fourth, and finally, if such unhappy events should come to pass, we would also
need to have the capability to reconstitute our nuclear forces above the levels which
we otherwise would plan to move to under the two START treaties, in order to
hedge against the possibility that such unfriendly regimes might seek to reassert
their military power by increasing their nuclear forces. Precisely because such sce-
narios are unattractive, we are putting significant stock in our efforts to strengthen
democratic governments, to accelerate START I deactivations even before the treaty
formally enters into force, and to achieve the earliest ratification of START I and
START II.
The Emerging Nuclear Threat
As Secretary Aspin recently reported to the House Armed Services Committee,
more than 20 other nations have embarked on efforts to develop weapons of mass
destruction. Regional powers continue to develop and obtain nuclear weapon tech-
nology and the means to deliver nuclear weapons. Although these regional powers
do not now pose a direct threat to the U.S., the danger to our national interests
and to our allies is growing. The proliferation of nuclear weapons is a major concern
to us as we conduct our Bottom-Up Review. It is also behind our efforts to put re-
newed emphasis into an across the board effort to stem the flow of nuclear weapons
technology and nuclear weapons-related technology; defining and implementing this
"counterproliferation initiative" will be one of the department's highest priorities.
Our forces must be also able to deter those regional powers who succeed in obtain-
ing nuclear weapons. We must maintain a viable nuclear force in order to deter re-
gional powers from threatening the use of weapons of mass destruction. Despite our
efforts to prevent proliferation, we must assume that some states will succeed in
their effort to obtain these weapons and may be willing to use them. The threat to
our allies and potentially to our own deployed forces would be significant, and to
prevent coercion on the part of a regional power will require a convincing nuclear
force. While this threat will not drive the configuration of U.S. strategic forces, our
strategic systems can play an important role in deterring the use of nuclear weap-
ons by a proliferant state.
Shaping our Strategic Forces
Shaping our nuclear forces involves many variables. We have always, however,
abided by one overriding principle — we require a fully credible force with a range
of capabilities. By this we mean truly a deterrent force, one that so complicates a
potential enemy's calculations that he would never consider initiating an attack.
Through the years the Triad has provided that credibility through a redundant and
complementary mix of weapons systems. Each has its own strengths and weak-
nesses, combining to form a deterrent which has eluded exploitation.
With the changing security environment, we have been able to eliminate those nu-
clear systems which are no longer required and to remove from alert that portion
of our bomber force, which throughout the Cold War was capable of being launched
on very short notice. The Minuteman II force was removed from alert status in Sep-
tember 1991, and we are currently removing all of the missiles from their launchers.
We have retired all nuclear-armed B-52Gs, and are in the process of retiring all
Poseidon submarines carrying the C-3 and C-4 missiles. President Bush's nuclear
initiatives of September 1991 and January 1992 resulted in reductions or termi-
nations to the B-2 bomber program, the Advanced Cruise Missile and the Short
Range Attack Missile II. These changes have not weakened our deterrent posture,
but reflect adaptations in our requirements to the changing security environment
and to the flexibility of our Triad posture. With START II, we look toward reducing
51
our deployed strategic forces further to 3500 warheads. The force that carries those
3500 deployed warheads must, however, be survivable, flexible, have off-setting
vulnerabilities, and be fully capable of carrying out national policy. A mix of Trident
submarines, single warhead Minuteman III missiles, and heavy bombers will meet
these requirements. We are currently examining what the exact make-up of the
force will be.
GAO Findings
The GAO report on the Triad had taken over 3 years to complete and raises many
good issues. Reports of this nature represent an extensive effort and are valuable
for us when reviewing our previous decisions. Many changes have occurred in our
forces since the work on of the report began, and many more are still to occur.
In analyzing the Triad, the GAO has understandably reduced this very complex
subject into smaller, more manageable portions, and focused on a more straight-
forward cost effectiveness analysis. However, with such an approach many of the
subtle relationships of the systems of the Triad are lost in what becomes a simple
"black and white" cost analysis. When the systems are compared on a cost-per-arriv-
ing warhead basis, as the GAO has done, the interactions and interdependencies
among the Triad legs become lost as subjective arguments that cannot be quantified.
The Triad's strength has always been derived from the ability of its various legs to
"fill the gap" of vulnerabilities of the others. It is not a question of which system
cost the least or is the least vulnerable to a particular threat that is important. The
important issue is whether the sum strength of the Triad has in fact caused such
insurmountable difficulties for the foreign adversary, that its deterrent effect has
been successful. I believe that for the U.S. strategic Triad, this is the case.
An example of the difficulty of completing an analysis of this nature over this long
a period of time is evident in the comparison of intelligence data. We take exception
to the GAO criticism that current U.S. strategic forces — as well as those being de-
ployed now — provide excessive capabilities, on the basis that the threat projections
made in the early 1980s did not materialize as predicted. The projections of the
early 1980s were grounded in a long history of ongoing Soviet force structure mod-
ernization; certainly no projection of any type could have envisioned the changes
which former President Gorbachev began during his presidency, to say nothing of
those which have occurred since August 1991.
We also do not accept the GAO's view that the stealth characteristics embodied
in the B— 2 and the Advanced Cruise Missile were not justified by the Soviet air de-
fense system. Modern Soviet fighter and surface to air missiles posed significant
threats to non stealthy systems. In addition, such systems are now found elsewhere
around the world. Our decision to develop low observable systems such as the B—
2 and the Advanced Cruise Missile reflected two sound judgments: (1) that we need-
ed to replace and fully modernize a critical leg of the Triad that was then — and is
still — underpinned by the increasingly limited capabilities of the B-52; and (2) that,
by developing stealth technologies, we were making an investment that played to
a long-term comparative advantage for the U.S. in its strategic competition with the
Soviet Union. The GAO report also omits any reference to the additional, force mul-
tiplying capabilities of the bomber force, i.e., its ability to fulfill both conventional
and nuclear missions.
Finally, I must note that despite all the work which went into it, the GAO re-
{>ort — with its focus on the interpretation of the world as we knew it even in the
ate 1980's — is not a paricularly good guide to force planning in the 1990's. While
some of the lessons GAO drew from its analysis of the Cold War nuclear situation
are still valid, many others do not accord with the new realities we face. It is to
these new realities that we must devote our attention as we consider the future of
U.S. strategic forces.
Mr. Chairman, this concludes my prepared remarks. I look forward to answering
whatever questions you might have and to providing more detailed information to
the Congress on the future shape of U.S. strategic forces when that information be-
comes available.
52
TEAM B:
THE TRILLION
Two experts report on how a group of Cold War true believers
were invited to second-guess the CIA. Did the "outside experts"
of the 1970s contribute to the military buildup of the 1980s?
Bv ANNE HESSING CAHN
Election years have much in common.
They produce a profusion of punditry,
media attention, and politically expedi-
ent action, quickly forgotten, and with
little lasting impact. But not always; some-
times events are set into motion that have
long lifetimes. This was the case in 1976 when,
as in 1992, an incumbent Republican president
faced a strong challenge from the right wing of
his own party. Then (as last year) sops were
offered to placate the far right and, while it is
too early to know which of the 1992 capers will
endure, we now know a great deal about one
of the most political events of 197G, and its re-
markably long-lasting effects on U.S. policy.
Late last year, the Central Intelligence
Agency (CIA) released the 1970 "Team B" re-
ports. Team B was an experiment in competi-
tive threat assessments approved by then-Di-
rector of Central Intelligence George Bush.
Teams of "outside experts" were to take inde-
pendent looks at the highly classified data
used by the intelligence community to assess
Soviet strategic forces in the yearly National
Intelligence Estimates (NIEs). NIEs are au-
thoritative and are widely circulated within
the government. U.S. national security policy
on various issues as well as the defense budget
are based on their general conclusions. Al-
(coiitimied mi /Kigc -j)
Anne Hessing Calm, a visiting scholar at the
Cmterfor Mental ioual Studies at the I hiiver-
sitii of Maryland in College Pari;, is a former
official at the U.S. Arm* Control and Disar-
mament Agency and the Defense Department.
22 The Bulletin of the Atomic Scientists
53
The Team B
experiment was
concocted by
conservative
cold warriors
determined to
bury detente.
though NIEs represent the collective judg-
ment of the entire intelligence community, the
lead agency is the CIA.
There were three "B" teams. One studied
Soviet low-altitude air defense capabilities,
one examined Soviet intercontinental ballistic
missile (ICBM) accuracy, and one investigated
Soviet strategic policy and objectives. But it is
the third team, chaired by Harvard professor
Richard Pipes, that ultimately received con-
siderable publicity and is commonly referred
to as Team B.
The Team B experiment was concocted by
conservative cold warriors determined to bury
detente and the SALT process. Panel mem-
bers were all hard-liners. The experiment was
leaked to the press in an unsuccessful attempt
at an "October surprise." But most important,
the Team B reports became the intellectual
foundation of "the window of vulnerability"
and of the massive arms buildup that began
toward the end of the Carter administration
and accelerated under President Reagan.
How did the Team B notion come about? In
1974, Albert Wohlstetter, u professor at the
University of Chicago, accused the CIA of sys-
tematically underestimating Soviet missile de-
ployment, and conservatives began a concert-
ed attack on the CIA's annual assessment of
the Soviet threat. This assessment— the
NIE — was an obvious target.
In the mid-1970s, the CIA was vulnerable
on three counts. First, it was still reeling from
the 1975 congressional hearings about covert
assassination attempts on foreign leaders and
other activities. Second, it was considered
"payback time" by hard-liners, who were still
smarting from the CIA's realistic assessments
during the Vietnam war years — assessments
that failed to see light at the end of the tunnel.
And finally, between 1973 and 1976. there were
four different directors of central intelligence,
in contrast to the more stately progression of
four directors in the preceding 20 years.
The vehicle chosen from within the adminis-
tration to challenge the CIA was the Presi-
dent's Foreign Intelligence Advisory Board
(PFIAB). Formed as the Board of Consultants
on Foreign Intelligence Affairs by President
Eisenhower in 1956, PFIAB was reconstitut-
ed by President Kennedy in 1961 after the
Bay of Pigs fiasco. Members are appointed by
the president but hold no other government
positions (except possibly on other advisory
committees or panels). By 1975. PFIAB was a
home for such conservatives as William Casey.
John Connally. John Foster. Clare Booth Luce,
and Edward Teller.
The PFIAB first raised the issue of compet-
itive threat assessments in 1975. but Director
of Central Intelligence William Colby was able
to ward them off, partly on procedural
grounds (an NIE was in progress). But Colby,
a career CIA officer, also said, "It is hard for
me to envisage how an ad hoc 'independent'
group of government and non-government an-
alyst* could prepare a more thorough, compre-
hensive assessment of Soviet strategic capa-
bilities— even in two specific areas — than the
intelligence community can prepare."1
At a September 1975 meeting of CIA, Na-
tional Security Council, and PFIAB staff, the
deputy for National Intelligence Officers,
George A. Carver, noted that since John Fos-
ter and Edward Teller, the principal PFIAB
members pushing for the alternative assess-
ment, disagreed with some of the judgments
made by the intelligence community, "the
PFIAB proposal could be construed as recom-
mending the establishment of another organi-
zation which might reach conclusions more
compatible with their thinking."
In 1976, when George Bush became the new
director of central intelligence, the PFIAB
lost no time in renewing its request for com-
petitive threat assessments. Although his top
analysts argued against such an undertaking.
Bush checked with the White House, obtained
an O.K., and by May 26 signed off on the ex-
periment with the notation, "Let her fly!! O.K
G.B."- Why in the world did the Ford adminis-
tration, gearing up for an election campaign,
put prominent outside critics of the CIA on
the agency's payroll, give them free access to
the classified material, data, and files they re-
quested, and not foresee how damaging the re-
sulting study could be?
B
y spring 1976, President Ford was in
deep political trouble. A January poll showed
that his performance had a 46 percent disap-
proval rating. The president attributed much
of the dissatisfaction to the increasing criti-
cism of detente by a conservative coalition in
both parties. Moreover, at the time the Soviet
Union and Cuba were actively supporting the
Popular Movement for the Liberation of An-
gola, while the U.S. Senate had barred fur-
ther covert American support to the other
contenders.
Nevertheless, early in January 1976 Presi-
dent Ford defended the policy of detente he
had inherited from Richard Nixon and said in
an S'BC News interview: "I think it would be
very unwise for a President— me or anyone
eis< — to abandon detente. I think detente is in
the best interest of this country. It is in the
best interest of world stability, world peace."
But then came the February 24 New Hannv
sliire primary, and President Ford nosed out
challenger Ronald Reagan by only one percent-
age point. Reagan began to step up his attacks
24 The Bulletin of the Atomic Scientists
54
on the "Ford-Kissinger" foreign )K>licy, claim-
ing that the Ignited States had lieen permitted
to slide into second place and that the Soviet
Union was taking advantage of detente at the
expense of American prestige and security.
In March, three important events took place.
During an interview, President Ford abruptly
banished the word "detente" from his political
vocabulary, much to the surprise of the White
House staff. "We are going to forget the use of
the word detente," the president said. "What
happens in the negotiations . . . are the things
that are of consequence.'" Then, at a lunch at
Washington D.C.'s Metropolitan Club, Richard
Allen, Max Kampelman, Paul Nitze, Eugene
Rostow, and Elmo Zumwalt, all well-known
hawks opposed to detente, agreed to form the
"Committee on the Present Danger" (CPD) to
alert the public to the "growing Soviet
threat." The first draft of the committee's ini-
tial statement was circulated to its members
within a month. Finally, on March 23. Ronald
Reagan won the North Carolina primary —
only the third time in U.S. history that a chal-
lenger had defeated an incumbent president in
a primary. He went on to win the Nebraska
and Texas primaries as well.
By now, conservative critics in full swing
kept up a steady cry of alarm. Paul Nitze. a
CPD and Team B member, testified before the
Joint Committee on Defense Production that
the Soviet Union was conducting a massive
civil defense program that would give it a bar-
gaining edge in the then-deadlocked arms
talks. Retired Defense Intelligence Agency
Director Lt. Gen. Daniel Graham, also a Team
B member, wrote in the September 1976
Reader's Digest "The Soviets have not built
up their forces, as we have, merely to deter a
nuclear war. They build their forces to fight a
nuclear war and [they] see an enormous per-
suasive power accruing to a nation which can
face the prospect of nuclear war with confi-
dence in its survival."
A January 21 . 1976. Library of Congress re-
port, "The U.SVSoviet Military Balance, a
Frame of Reference for Congress," identified
a strong shift in the quantitative military bal-
ance toward the Soviet Union over the past 10
years. And the CIA itself revised its estimate
of Soviet military spending to 10-15 percent of
Sonet gross national product (GNP). as com-
pared to 6-S percent in previous NIEs. The
revision was immediate news.
(This jump did not indicate any great in-
crease in Soviet military spending nor did il
change the Pentagon's estimates of actual So-
viet troops, tanks, and missiles. Indeed, it re-
flected the judgment that the Soviet military
sector was less efficient than previously be-
lieved and therefore the military's economic
burden on the Soviet Union was greater than
earlier estimates indicated. None of this
meant a greater threat to the United Slates.
However, such distinctions, usually made in
the next to last paragraph of a long article,
were lost on the public, and the message
seemed to be that the Russians were spending
more on defense and therefore we should too.)
In the summer of 1976, President Ford was
rearranging priorities in much the same errat-
ic way as George Bush did 16 years later in an
effort to stave off conservative critics. Even
the signing of the Peaceful Nuclear Explosions
Treaty was delayed from May 12 to May 28
because of panic at Ford's loss to Ronald Rea-
gan in the Nebraska primary.
After Carter
was elected,
George Bush
made sure
that Team B
became front-
page news.
I
n July 1976. Director of Central Intelli-
gence George Bush let a PF1AB subcommit-
tee suggest members of the three B teams; in
August he wrote to the president that "morale
at the CI A«s improving."'
Each K team met in September and October
and exchanged drafts with their CIA counter-
parts during October. The first press leak oc-
curred two days after the first meeting of the
CIA and Team B members who were examin-
ing Soviet strategic policy and objectives.
William Beecher's story in the October 20
Boston Globe contained leaks by at least one
Team B member who conveyed to the journal-
ist only his recommendations, not those of his
fellow panelists. According to Leo Cherne,
then chairman of PFIAB, Director of Central
Intelligence Bush was aghast at the leak and
stormed into the Old Executive Building ac-
cusing members of PFIAB of being the leak-
ers. Cherne assured Bush that this was not
the case, and that "members of PFIAB were
sufficiently smart to recognize that any public-
ity would invalidate what had been a serious
effort.'" The story was not picked up and
seemed to fade from view.
However, after the Democrats won the elec-
tion and President-elect Jimmy Carte)- had ig-
nored Bush's hint that up to now. CIA direc-
tors had not changed with an incoming admin-
istration. George Bush, the foe of leaks,
agreed to meet with David Binder of the S'eiv
York Times. The same director who wrote to
President Ford in August 1976. "1 want to get
the CIA off the front pages and at some point
out of the papers altogether." now made sure
that Team B would become front-page news/
On Sunday. December 26. the lead Srir
York Times story was about Team I>. Bush ap-
peared on Mat tin I'rcxx, and three separate
congressional committees vowed to hold hear-
ings on the whole exercise. Although officials
within the new Carter administration paid
scant attention to the Team B reports, the
April ]!«>:; 25
55
Simultaneously,
the newly
formed
Committee on
the Present
Danger i
hammered the
message home.
spadework had been done. In particular, the
Pipes panel's major conclusions had been pub-
licly and repeatedly aired.
Meanwhile, back in November, nine days
after the presidential election, the Committee
on the Present Danger issued its founding
statement, "Common Sense and the Common
Danger." "The principal threat to our nation,
to world peace and to the cause of human free-
dom is the Soviet drive for dominance based
upon an unparalleled military buildup The
Soviet Union has not altered its long held goal
of a world dominated from a single center —
Moscow." If this sounded similar to the conclu-
sions of Richard Pipes 's Team B panel, it was
hardly surprising; panel members Paul Nitze,
Richard Pipes, and William Van Cleave had
leading roles in the committee.
Even before the Team B report was officially
presented to PFIAB, Pipes was eager to publi-
cize its findings. He opened a December 7
meeting by discussing the possibility of declas-
sifying the report. After the CIA rejected de-
classification, Pipes said that "he would urge
PFIAB to make the Team B report available to
as large an audience as possible. If his appeal
to PFIAB were rejected ... he mentioned . . .
the publication of articles on the general sub-
ject of the report without reference to classi-
fied information Pipes also raised the possi-
bility of using the Freedom of Information Act
to get the report into the public domain."*
I
t took 16 years before Pipes's hopes were
fully realized and the documents published. In
February 1989, 1 filed a Freedom of Informa-
tion Act request to obtain Team B documents.
After repeated letters, phone calls, and an in-
terview by the chairman of the Intelligence
Council produced only two items, I filed a
complaint in the U.S. District Court in July
1992. By the first meeting before the judge
in September 1992, counsel for the CIA
promised that I would receive all the docu-
ments before the end of October. The CIA de-
posited the Team B report at the National
Archive, and delivered to me most of the docu-
ments I had requested before the end of Octo-
ber 1992.
Today, the Team B reports recall the stri-
dency and militancy of the conservatives in
the 1970s. Team B accused the CIA of consis-
tently underestimating the "intensity, scope,
and implicit threat" )x>sed by the Soviet Union
by relying on technical or "hard" data rather
than "contemplat[ing] Soviet strategic objec-
tives in terms of the Soviet conception of
'strategy' as well as in light of Soviet history,
the structure of Soviet society, and the pro-
nouncements of Soviet leaders."
And when Team B looked at "hard" data,
everywhere it saw the worst case. It reported,
for instance, that the Backfire bomber "proba-
bly will be produced in substantial numbers,
with perhaps 600 aircraft off the line by early
1984." (In fact, the Soviets had 235 in 1984.)
Team B also regarded Soviet defenses with
alarm. "Mobile ABM [anti-ballistic missiles]
system components combined with the de-
ployed SAM [surface-to-air missile] system
could produce a significant ABM capability."
But that never occurred.
Team B found the Soviet Union immune
from Murphy's law. They examined ABM and
directed energy research, and said, "Under-
standing that there are differing evaluations
of the potentialities of laser and CPB [charged
particle beam] for ABM, it is still clear that
the Soviets have mounted ABM efforts in both
areas of a magnitude that it is difficult to
overestimate. " (Emphasis in original.)
But overestimate they did. A facility at the
Soviet Union's nuclear test range in Semi-
palatinsk was touted by Gen. George Keegan,
Chief of Air Force Intelligence (and a Team B
briefer), as a site for tests of Soviet nuclear-
powered beam weapons. In fact, it was used to
test nuclear-powered rocket engines. Accord-
ing to a Los Alamos physicist who recently
toured Russian directed-energy facilities, "We
had overestimated both their capability and
their [technical] understanding."
Team B's failure to find a Soviet non-acous-
tic anti-submarine system was evidence that
there could well be one. "The implication could
be that the Soviets have, in fact, deployed
some operational non-acoustic systems and
will deploy more in the next few years." It
wasn't a question of if the Russians were com-
ing. They were here. (And probably working
at the CIA!)
When Team B looked at the "soft" data con-
cerning Soviet strategic concepts, they slanted
the evidence to support their conclusions. In
asserting that "Russian, and especially Soviet
political and military theories are distinctly of-
fensive in character," Team B claimed "their
ideal is the 'science of conquest' (nauka
pobezltdat) formulated by the eighteenth-cen-
tury Russian commander. Field Marshal A.V.
Suvorov in a treatise of the same name, which
has been a standard text of Imperial as well as
Soviet military science." Raymond Garthoff, a
senior fellow at the Brookings Institution, has
pointed out that the correct translation of
nauka jmbezlidat is "the science of winning" or
the "science of victory." All military strate-
gists strive for a winning strategy. Our own
military writings are devoted to winning vic-
tories, but this is not commonly viewed as a
policy of conquest.
Team B hurled another brickbat: the CIA
26 The Bulletin of the Atomic Scientists
56
consistently underestimated Soviet military
expenditures. With the advantage of hind-
sight, we now know that Soviet military
spending increases began to slow down pre-
cisely as Team B was writing about "an in-
tense military buildup in nuclear as well as
conventional forces of all sorte, not moderated
either by the West's self-imposed restraints or
by SALT." In 1983, then-deputy director of
the CIA, Robert Gates, testified; The rate of
growth of overall defense costs is lower be-
cause procurement of military hardware — the
largest category of defense spending — was al-
most flat in 1976-1981 . . . [and that trend] ap-
pears to have continued also in 1982 and 1983."
While Team B waxed eloquent about "con-
ceptual failures," it was unable to grasp how
the future might differ from the past. In 197G
mortality rates were rising for the entire Sovi-
et population, and life expectancies, numbers
of new labor entrants, and agricultural output
were all declining. Yet Team B wrote confi-
dently. "Within what is, after all, a large and
expanding GNP . . . Soviet strategic force* have
yet to reflect any constraining effect of civil
economy competition, and are unlikely to do
so in the foreseeable future." (Emphasis in
original.) And When Ronald Reagan got elect-
ed, Team B became, in essence, the "A Team."
For more than a third of a century, percep-
tions about U.S. national security were col-
ored by the view that the Soviet Union was on
the road to military superiority over the Unit-
ed States. Neither Team B nor the multibillion
dollar intelligence agencies could see that the
Soviet Union was dissolving from within.
For more than a third of a century, asser-
tions of Soviet superiority created calls for the
United States to "rearm." In the 1980s, the
call was heeded so thoroughly that the United
States embarked on a trillion-dollar defense
buildup. As a result, the country neglected its
schools, cities, roads and bridges, and health
care system. From the world's greatest credi-
tor nation, the United States became the
world's greatest debtor — in order to pay for
arms to counter the threat of a nation that was
collapsing. ■
1. William E. Colby to President Ford (Nov. 21,
1975), author collection. Obtained under the Freedom
of Information Act by Anne H. Cahn.
2. George A. Carver. Jr.. "Note for the Director." Ma\
26, 1970.
3. Raymond Garthoff. Dtteiite and Ciiufrtiiitalimi:
American. Sorhi Relutioa* From Xisw, it, Hintian
(Washington. D.C.: Brookings Institution. liiV.'. p. :Ms
4. Ibid.
5. Director of Central Intelligence George Hush to
President Ford (August .:. IHTIil. author collection.
fi. Leo Cherne. interviews with author May 2.I. 1!)!KI:
August 2. HUM.
7. Leo Cherne. May 2.:, HUM.
S. Donald Suda. note to file (December 7. 1!)7U). au-
thor collection.
57
DOLLAR
EXPERIMENT
ByJOHNPRADOS
By the mid-1 970s, millions of Americans had be-
come disillusioned. A president had resigned in
disgrace. Saigon had fallen. And the Nixon-Kiss-
inger policy of detente with the Soviets was widely
suspect The time was ripe for a major initiative
by conservatives. Among the key players on
"Team B" and the Committee on the Present Dan-
ger were diplomat Paul Nitze (upper left), Richard
Pipes (left), and scientist Edward Teller (above).
Remember the "window of vulnerabili-
ty?" Russians, ten feet tall, using in-
credibly accurate intercontinental bal-
listic missiles, would be able to elimi-
nate America's nuclear deterrent with only a
fraction of their strength. Then, remaining So-
viet nuclear forces would threaten to vaporize
American cities, paralyzing Washington. Mos-
cow could then have its way.
These hy]>othetical Soviet nuclear capabili-
ties were expected to peak in the early
1980s— the "window of vulnerability." As used
in Ronald Reagan's 1980 presidential cam-
paign, that phrase conveyed the specter of a
Soviet diplomatic offensive' the like of which
had never been seen. To close the window as
president. Reagan embarked on a military
spending spree that, by itself, went a good dis-
tance toward plunging the United States into
the economic crisis we face today. The pur-
ported threat never materialized.
Originally the threat had been delineated in
an intelligence report and in a top-secret ex-
periment in threat analysis. Those who never
saw the secret reports or the leaks had the
message drummed into them by a conserva-
tive organization, the Committee on the Pre-
sent Danger, which actively opposed Carter
administration policies. Many of the commit-
tee's key members went on to top posts in
Ronald Reagan's Washington. Tints, in certain
fcoiitiimvd on jxif/r J?)
.Join/ I'railos, a historian ol national xccHi'ittf
based in Waslihifftuv. D.C.. is the author of
Keepers of the Keys: The National Security
Council from Truman to Bush (l!HU).
April 19WI 23
58
— — "Piffy-ab"
PRADOS (runt.fnnv p. j.i) members
crucial ways, a mid-1970s intelligence "experi- tugyp annOVed
ment" proved central to the American experi- WCI c au *
ence of the 1980s. that CIA
It is now possible to revisit this formative .
episode, thanks in part to the efforts of author analyses Were
Anne Cahn (see page 22), who pressed for the jjqJ ^ douf aS
declassification of the documents. On Septem-
ber 16, 1992, the CIA declassified its 1976 Na- their 0WI1.
tional Intelligence Estimate 11-3-/8-76 (1976-
NIE), titled "Soviet Forces for Intercontinen-
tal Conflict Through the Mid-1980s," and the
related "Intelligence Community Experiment
in Competitive Analysis: Soviet Strategic Ob-
jectives, An Alternative View."
National intelligence estimates (NIEs) pre-
sent the considered view of the U.S. intelli-
gence community on vital subjects or prob-
lems. NIEs are official publications of the Di-
rector of Central Intelligence (DCI), the se-
nior intelligence officer of the U.S. govern-
ment. Rut the intelligence experiment con-
ducted in 1976 included a so-called "competi-
tive analysis," in which the official analysts'
work was to be matched against that of a
group of outside consultants called "Team B."
The result proved devastating.
S
_'pecific motivation for the experiment
flowed from the President's Foreign Intelli-
gence Advisory Board (PFIAB), a White
House watchdog unit set up by President
Dwight D. Eisenhower to monitor the intelli-
gence community. The board (the "Piffy-ab")
traditionally reported to the president twice a
year on general intelligence matters, and it
also did postmortems on important events,
such as the Cuban Missile Crisis, the Tel of-
fensive, or aircraft over-flight incidents. Since
1969. PF1 AB had supplied presidents with an-
nual surveys of Soviet power separate from
the national estimates. Some board members
were annoyed that CIA analyses were not as
dour as their own. In 1975. and again the fol-
lowing year, PFIAB recommended a competi-
tive analysis. When first suggested. Director
of Central Intelligence William Colby resisted
the proposed experiment, but in 1976 the DCI
was George Bush, who readily cooperated
with the White House and PFIAB officials in
selecting consultants for the exercise and ex-
changing letters with PFIAB to set the
ground rule.- for the exercise.
A politician to his fingertips. Bush was as
political as DCI as he would he as president
years later, although he had told a confirma-
tion hearing that he would avoid politics. With
the Team B exercise already in motion, he de-
April l *>!»:: 27
59
Team B did not
actually analyze
data; it relied
on ideological
judgments . . .
clared to the Joint Economic Committee, "I
think we have done . . . whiit is essential to see
that estimates are protected from policy
bias.'" The opposite turned out to be the case.
Selected as team leader for the experiment
was Harvard historian Richard Pipes, a con-
servative who had just edited a collection of
papers on Soviet strategy in Europe. (Until
that time his work had focused primarily on
Russian pre- and early post-revolutionary his-
tory.) Other members were also conserva-
tives!— Dr. William R. Van Cleave, professor at
the University of Southern California; retired
Gen. Daniel 0. Graham, a former head of the
Defense Intelligence Agency; retired Col.
Thomas W. Wolfe, a figure in air force intelli-
gence; and a retired air force chief, John Vogt.
There was also an advisory panel that includ-
ed Amb. Foy Kohler, Paul Nitze, Seymour
Weiss, air force Gen. Jasper Welch, and Paul
Wolfowitz — all outspoken conservatives.
One man's experiment became another's op-
portunity. The competition was supposed to lie
in the fact that two teams with different views
would use the same set of data to produce anal-
yses. In actuality, Team B took the exercise as
a mandate to deconstruct CIA analyses, even
criticizing estimates cast decades earlier.
Team B's report was not an alternative NIE
at all. Instead, it started with the judgments
about Soviet strategic objectives underlying
the NIEs, continued with critiques of NIE
judgments on substantive issues, then re-
turned to Soviet objectives for more criticism.
When Team B members met with the ana-
lysts who worked on the official NIE to hold a
debate for the benefit of senior officials, the
exchanges proved bloody. Howard Stoertz.
the agency's national intelligence officer for
the Soviet Union, who led the CIA group for-
mally responsible for the NIE, was unable to
stem the intellectual rout. "It was an absolute
disaster for the CIA." one official recounted.1'
w
I hat was it about Team K's arguments
that carried such force? The recently declassi-
fied Team B report itself reveals nothing stun-
ning at all. The criticisms leveled at the CIA
analysts were a collection of standard cautions
(with illustrations from past NIEs). plus gen-
erous dollops of conservative presumption. It
was the kind of thing (minus the ideological
bent) that was often served up to analysis in
training. But this time it was used in a back-
alley cat fight in which Team B members of
lowering reputation overawed respectful
agency counterparts.
For example, the Team 1! critique held that
the NIEs were flawed because they concen-
trated on hard data! And NIE authors were
declared guilty of "misinterpreting" softer in-
formation such as Soviet writings on military
doctrine.1
What were the NIEs missing? Team B ar-
gued that the estimates ignored "the possibili-
ty that the Russians may be pursuing not a
defensive but an offensive strategy."' (Original
in italics.) The CIA was castigated for pre-
senting its assessments without evidence:
that Moscow was interested in capping the
arms race; that the Soviets sought parity;
that Russians feared encroachment due to
past invasions and wars; and that Soviet lead-
ers might wish to limit military spending in
order to shift resources to the civilian sector.
According to Team B, these things were mere
suppositions.
As Team B's report put it (using italic type
for emphasis): "The NIEs are filled with un-
supported and questionable judgments about
what it is that the Soviet government wants
and intends. It is this practice, rather than the
lack of solid information, that has caused in the
past (and in considerable measure does so in
the present) recurrent underestimations of
the intensity, scope, and implicit threat of the
Soviet strategic buildup."'
Of course, Team B's contention that Soviet
thinking was inherently offensive was equally
unsupported by "solid information." Team B
simply asserted that Russia had not suffered
"an exceptional number of invasions and inter-
ventions" (obviously a question of interpreta-
tion).' Team B's other evidence included state-
ments about the importance of forward move-
ment in Soviet political theory, and "the lack of
any kind of genuine legitimacy on the part of
the Soviet government [which] compels it to
create its own pseudo-legitimacy.'" This rea-
soning was not grounded in data. Team B
members merely picked out their favorite quo-
tations from Marxist-Leninist writings, exact-
ly as did authors for Pravda.
An especially pernicious Team B contribu-
tion was the description of Soviet thinking as
"Clausewitzian." Using the term tended to
brand Soviel policy as somehow militaristic by
definition, as if American and allied generals
did not read and use the ideas of the noted
Prussian military philosopher. Even ideas
were guilty by association, and the CIA's 1976
analysts were guilty of the sins of drafters of
previous national estimates.
The substantive disputes raised by the re-
port for the most part concerned long-stand-
ing intelligence issues over which analysts dif-
fered. Team I', members merely took views
other than those expressed in the NIEs. By
including these disputes. Team B ap)>eared to
lend substance to the group's commentary, but
the report was actually a more hortatory
analysis of objectives.
28 The Bulletin of the Atomic Scientists
60
Some disputes were dredged up from the
past— like the rate of Soviet 1CBM deploy-
ments, or the anti-missile potential of Soviet
anti-aircraft missiles.' In other areas, like the
debate over the capabilities of the Soviet
Backfire bomber, the NIEs were held to be in-
accurate, although no one knew the truth, in-
cluding Team B. (The Team B report took the
quaint position that the existence of the Soviet
Strategic Rocket Forces as a separate military
service was an "administrative device" unre-
lated to the importance Moscow placed on con-
trolling nuclear weapons.)
I
This was a classic dilemma. Was the Soviet
glass half empty or half full? It was Khrush-
chev himself who had coined the phrase
"metal eaters," and Westerners could find evi-
dence of the strain on the civilian economy
every day on the streets of Moscow, where
Russians' desperate desire for consumer
goods could only be termed obvious. Never-
theless, Team B helped to redefine the terms
of the debate in such a way that through the
remaining 1970s and the 1980s the Soviet
... and it was
critical of the
CIA for relying
on data.
t is not possible to review in detail all the is-
sues raised by Team B in a brief article. But
one deserves mention, because it has been
central to the entire outcome of the Cold
War— the estimates of Soviet military spend-
ing. Soviet spending was one of Team B mem-
ber Daniel Graham's hobby-horses and. not
surprisingly, the issue was taken up in the
team's substantive critique.
At the time, it was fashionable to criticize
the CIA's building-block model of the Soviet
economy as understating the level of military
spending. And need for improvement had be-
come evident when U.S. intelligence gained
access to a copy of a Soviet budget document
that showed the Russians' own secret projec-
tion of their military spending levels— which
were roughly twice what the CIA had esti-
mated. Official assessments of Soviet military
spending were revised to 11-13 percent of
gross national product (GNP) rather than the
6-8 percent the CIA had earlier estimated.
So far. so good. But the problem was. how
great a burden did such spending place on the
Soviet economy? Given their ideological bent.
Team B argued that the Soviets had no con-
flicts over resource allocation, except perhaps
between military services. The Soviet Union
would spend whatever was necessary to attain
strategic superiority and a war-winning capa-
bility against the United States.
As Team B put it: "There have been in tlie
past some disputes between heavy industry
and medium and light industry sectors of the
Soviet economy. The heavy industry spokes-
men (from time to time referred to as 'metal
eaters') include the producers of military
equipment, but this [these disputes] is not
convincing evidence of a civilian economy chal-
lenge to military resource allocation priority."
Team B attacked several NIEs for suggesting
that economic considerations might restrain
Soviet military growth. Nikita Khrushchev's
effort to reform the Soviet military and
limit spending was not classed as convincing
evidence.
The real danger
As the Committee on the Present Danger began to press the
idea that the United States was falling behind the Soviet Union
in producing weapons of mass destruction and heading toward
vulnerability to nuclear blackmail, Bernard T. Feld, then edi-
tor-in-chief of the Bulletin, expressed his exasperation. (Feld,
remembered fondly at the magazine as Bernie, remained a
member of the Bulletin's Board of Directors until his death in
February 1993.)
"The country," he wrote in the February 1978 Bulletin, "is
being blanketed by one of the most insidious campaigns in re-
cent history; an attempt to turn back the clock to the worst
days of the Cold War."
Bernie argued that despite its saber-rattling, the Committee
on the Present Danger could not hope to convince the public
that the Soviet Union was already stronger than the United
States. "It is demonstrably clear (to all but far-out hawks and
ignoramuses) that ... the American nuclear arsenal is today
larger, more versatile, more lethal, and less vulnerable to a pre-
emptive first strike than that of the Russians."
But the committee's argument— that defense spending
should be increased to counter the aggressive intentions the
Soviets would show in the mid-1980s— was a more difficult
problem. The claim may have been fortune-telung, not analysis,
but the committee could always bolster its case with "readily
available quotes from blustering Russian generals and other
hard-liners."
And Bernie came closer than most to realizing the trillion dol-
lar implications of the committee's campaign: "What is at stake
is not simply a few billion dollars here or there for 'a moderate
program of political action and military improvement.' What is
at stake is a complete reversal of policy, an attempt to base
American (and, indeed, world) security on the open-ended pur-
suit of more and more lethal arms, the rejection of the principle
that we are all better off in a world of limited and controlled
arms rather than in an uncontrolled arms race.
"The present danger is not that we are failing prudently to
waste a few billion dollars and to keep up with Ivanovich in a
meaningless but benign nuclear arms competition. The danger
is that we are being convinced by our hawks to launch into a
new and qualitative spiral in the nuclear aims accumulation, to
reject the on-going mild, even somewhat inadequate, attempts
to find a roughly equitable ceiling at which to call a hall
"The present danger is not to realize that the Committee in
the Present Danger."
April IMS 29
61
Retired Gen. Daniel 0. Graham, whose "High Frontier' proposal foreshadowed
SDI, and John Foster, Jr., head of Lawrence Livermore National Laboratory,
("the most enthusiastic designer of nuclear weapons" in America, according to
Britain's Lord Zuckerman), were Team 6 members. Gov. Ronald Reagan of
California, who challenged President Gerald Ford in the 1 976 Republican
primaries, was an early Committee on the Present Danger enthusiast.
glass would only be seen as half full. Soviet
military power, according to the series of Pen-
tagon publications by that title, was truly
colossal, justifying a $1.5 trillion dollar in-
crease in U.S. military spending during the
Reagan administration.
B
'ut the Soviet colossus had feet of clay. In
the mid-1980s, the intelligence community qui-
etly halved their estimates of the accuracy of
the most dangerous type of Soviet ICBMs,
and with that, the window of vulnerability in-
stantly disappeared. In the late 1980s, military
authorities described Soviet military spending
as consuming 15-17 percent of GNP without
achieving military superiority. It became
harder to believe in the efficient, inexorable
drive for Soviet strategic dominance. Also in
the 1990s, Mikhail Gorbachev spoke of the ne-
cessity of reform in the Soviet economy,
agreed to a series of amis control proposals,
then unilaterally declared military with-
drawals from Afghanistan and Eastern Eu-
rope— unprecedented Soviet actions. Gor-
bachev's reforms were too little to ease Soviet
economic difficulties, however, and the gov-
ernment co'lapsed.
In short, the burden of Soviet military
spending, which was much greater than U.S.
intelligence projected (and completely dis-
counted by Team B) caused such dislocation in
the Russian economy that it brought about the
disintegration of the Soviet Union.
We now know that the military burden was
in play by the 1970s, probably already by the
late 1960s, and that the Soviet economy was al-
ready contracting in the Brezhnev era. It was
not necessary for the United States to spend
itself into penury' to
bring about these re-
sults. The worst aspect
of the Team B episode
was that it was used to
justify parallel U.S. ex-
penditures by exagger-
ating the Soviet threat.
Ironically, the justifi-
cation for increased de-
fense spending lay not
in the Team B report
itself, but in how the
Team B process influ-
enced the NIEs. The
experiment emboldened
those in the intelligence community who dis-
agreed with the national estimates. The
1976 NIE is especially enlightening in this
respect.
In the intelligence business, to dissent from
an NIE is to "take a footnote," although by
1976 "footnotes" were actually interpolations
presented in italics in the main text. A high
proportion of the 1976 NIE is composed of
footnotes, many of them from air force intelli-
gence chief George J. Keegan, but also from
the intelligence chiefs of the army and navy
and from the Defense Intelligence Agency,
then under Lieut. Gen. Samuel V. Wilson.
Keegan claimed in the 1976 NIE that the
record of footnotes in that document and in
previous NIEs represented the true history of
Soviet military power, with the majority opin-
ion in the NIEs falling "far short of grasping
the essential realities of Soviet conflict pur-
pose and evolving capability."lli This dissent
came despite the fact that, in a process of self-
censorship, the final "consensus" view in the
NIE was: "In our view, the Soviets are striv-
ing to achieve war-fighting and war-survival
capabilities which would leave the USSR in a
better position than the US if war occurred.
The Soviets also aim for intercontinental
forces which have visible and therefore politi-
cally useful advantages over the US. They
hope that their capabilities for intercontinen-
tal conflict will give them more latitude than
they have had in the past for the vigorous pur-
suit of foreign policy objectives, and that these
capabilities will discourage the US and others
from using force or the threat of force to influ-
ence Soviet actions.""
This NIE vision — while not ominous enough
to satisfy George Keegan — was nevertheless
so somber that it moved the director of the
State Department's Bureau of Intelligence
and Research to insert a footnote asserting
that the threat of Soviet behavior had been
exaggerated.
In fact, the effort to keep the air force and
other dissenters on board skewed the 1976
30 The Bulletin of the Atomic Scientists
62
NIE no much that the CIA took footnotes to
dissent from its own NIE — on the Backfire
bomber, on the range of the Soviet SS-20 mo-
bile missile, on the purpose of several large
radars, and on aspects of Soviet air defenses.
There may be other places where the CIA dis-
sented in portions of the NIE narrative that
remain classified.
A. he net effect of the 1976 NIE was to move
the intelligence community far down the road
to regarding Moscow as a near-term threat. If
that was not enough, key members of Team B,
many of whom became members of the Com-
mittee on the Present Danger, went public
with their version of the threat. Within
months, Richard Pipes produced an article ti-
tled "Why the Soviet Union Thinks It Could
Fight and Win a Nuclear War."11 Meanwhile,
Paul Nitze's articles and press briefings were
a major thorn in the side of the Carter admin-
istration, which was striving to achieve arms
control agreements with the Soviets. Pipes
later joined the Reagan administration's Na-
tional Security Council staff as its first direc-
tor for Soviet Affairs. Nitze and many others
from the Committee on the Present Danger
served in the Reagan administration, too.
George Bush, if one credits his pubb'c and
private expressions, took the position that
there was nothing special about the Team B
experiment or its outcome, except the leaks
that resulted. In his autobiography, Bush
writes that the most important intelligence es-
timate he produced as DCI concerned Leb-
anon. He does not mention Team B at all. On a
television news show (Face tiie Nation. Jan-
uary 2, 1977), Bush refused to be drawn out on
the substance of the dispute, but he said that
the Team B experiment had been useful, and
that he might recommend more. Privately, in a
secret memorandum to recipients of the na-
tional estimate, Bush wrote that its judgments
1. U.S. Congress. Joint Economic Committee, Hear-
ings: Allocation of liatoiircc* in the Sonet Union anil
China, 94th Cong.. 2nd sess. (Washington: Government
Printing Office, 1976), p. 61.
2. David Callahan, Dangerous Capabilities: Paul
NiUe and Oie Cold War (New York: HarperCollins,
1990) p. 380.
3. Central Intelligence Agency (CIA). Intelligence
Community Experiment in Competitive Analysis: So-
viet Strategic Objectives. An Alternate War: Report of
Team "B,~ December 1976 (declassified September 10.
1992). p. 1. Obtained under the Freedom of Information
Act by Anne H. Calm
4. CIA. Intelligence Community Experiment. |>. 12.
5. Ibid., p. 10.
6. Ibid., p. 14.
7. Ibid.
8. See John Prados. The Soviet Estimate: VS. Intel-
ligence Analysis and Soviet Strategic Forces (Prince-
ton: Princeton University Press, 1986). This work cov-
were arrived at after the |>arties "had the lioti-
efit of alternative views" but that "there is no
truth to . . . allegations" that the NIE had
been influenced by pressure from Team B.'
Readers may judge for themselves.
In 1978, the staff of the Senate Intelligence
Committee reviewed the Team B episode.14 It,
too, tried to minimize the damage, construing
the leaks as the problem, not the substance or
methodology of the experiment.
Michael Howard, a British historian, took a
more forthright position, perhaps because he
was not in government. His comments on the
Team B episode have survived the tests of
time and declassification. Speaking to the New
York Arms Control Seminar on March 31,
1977, Howard said that interpreting Soviet in-
tentions was like reading a Rorschach ink
blot — the act says more about the analyst than
it does about Soviet intentions.
Decision-making. Howard maintained, had
to be at least as confused in Moscow as in
Washington. Moreover, as Howartl recalled,
the same kinds of scare tactics were used U>
promote notions of a "Bomber Gap" and a
"Missile Gap" in the 1950s. The Soviet military
buildup could be explained in less convoluted
ways than an attempt to wrest superiority and
expand by force. The precepts of Marxist doc-
trine, he said, preached more peaceful expan-
sion. He drew an analogy between Russia and
Wilhelmine Germany — countries that were
psychologically insecure, feared encirclement,
and were conscious of backwardness. The de-
scription fits Brezhnev's Soviet Union in 1977.
But Team B had its way, and the Commit-
tee on the Present Danger could hardly have
been more pleased at the fashion in which
American politics evolved over the following
decade. It is unfortunate that today, with the
Cold War over, Americans are still paying the
price for the extravagance of the security de-
bate of the 1970s. It is doubly unfortunate
that our children, and even their children, will
still be paying. ■
ers these and other intelligence disputes in consider-
able detail.
9. CIA. Intelligence Cnmmuuitu Experiment, p. 10.
10. Central intelligence Agency. Soviet Fortes for
Intercontinental Conflict Through Uie Mid lDHOs. No-
tional Intelligence Estimate 11-3/8-7G, December 21,
1970 (declassified September 16, 1992), p. 6.
11. Ibid., p. 3
12. Richard Pipes, "Why the Soviet Union Thinks It
Could Fight and Win a Nuclear War." Commentary
(July l977).pp.21-»4.
IK. Office of the Director of Central Intelligence.
Mi-limrtiiidiiiit fur licciptrnts til .Xittiutittt Intelligence
Estimate lJ-J/X-Ttl. (late not available (declassified
Septemlier 10. 1992), p. 1.
14. Semite Select Committee on Intelligence. Report:
Tin Siitiitual intelligence Estimate* A-B Team
EpijttMli Concerning Soviet Strategic Ca/iabilttti and
Objectim. 95th Cong., 2nd sess. (Washington: Govern-
ment Printing Office. 1978).
The CIA's
own estimate
became so dark
that the CIA
began to insert
footnotes
taking issue
with its own
report
April UNCI 31
63
Senator
John Glenn
News Release
Tor Immediate Release: Contact: Bryan McCleary (202) 224-9799
June 28, 1993 Len Weiss (202) 224-4751
GLENN: "AIR FORCE LIED TO CONGRESS"
Senator Calls on Pentagon to Punish Those Responsible
Senator John Glenn (D-Ohio) is demanding the Pentagon launch
an investigation to determine how and why false data was given to
Glenn and congressional auditors regarding the ability of the B-
1B bomber to evade radar.
"I was lied to and I resent it," said Glenn, a supporter of
the B-1B on the Armed Services Committee. "I want to see those
responsible punished for their actions."
The call for an investigation comes in light of new evidence
provided to Glenn by the General Accounting Office (GAO) . Glenn
received the information in response to questions he raised in a
hearing before his Governmental Affairs Committee June 10.
A de-classified 1989 Pentagon memo given to Glenn reveals
that Department of Defense officials deliberately provided
congressional investigators with false B-l radar detection data
and withheld the correct information. This same faulty
information was again used in a letter to Glenn last year touting
the benefits of the B-1B. It was also given out to Congressional
investigators in 1991.
At issue is the B-lB's "radar cross section" which is the
profile that will appear of the plane at various angles when it
appears on enemy radar. The smaller the cross section, the
better the chance that the plane will not be picked up by enemy
radar. The size of the cross section was altered in the
documents given to Congress to make the plane's ability to evade
radar appear greater than it really was.
Glenn said the "credibility problem created by this
misinformation creates questions about other Air Force data --
including that for programs like the B-2."
Last year, the Air Force designated the B-1B as the backbone
in the Air Force's conventional bomber force -- an action which
Glenn agrees with. "Despite the controversy over radar cross
section, the B-1B is still a major improvement over the B-52, and
can be employed at a fraction of the cost of the B^,2," said
Glenn. "I'm angry, though, that the Pentagon has put me in this
difficult position," he added.
Glenn said he will raise this matter in a hearing Tuesday on
Air Force bomber issues before a Senate Armed Services
subcommittee. Scheduled to testify is Air Force General John Loh
who is responsible for the operational employment of virtually
all Air Force combat aircraft. General Loh wrote a 1992 letter
to Glenn citing the faulty data. (HEARING TUESDAY AT 9:30 a.m.
222 RUSSELL BUILDING)
See attached Glenn letter.
64
GAO
United State*
General Accounting Office
Washington, D.C. 20648
Program Evaluation and
Methodology Division
June 16, 1993
The Honorable John Glenn
Chairman, Committee on
Governmental Affairs
United States Senate
Dear Mr. Chairman:
This letter follows up on the Senate Governmental Affairs
Committee's hearing last Thursday (June 10) on GAO's
Evaluation of the U.S. Strategic Modernization Program.
As you will recall, in answer to your question about any
limitations of our study, I mentioned four kinds of data
problems that troubled me with regard to both the quality of
data being produced by the Department of Defense, and GAO's
access to data. The four problems were (a) data that were
themselves incomplete or of poor quality; (b) data from which
we were excluded via classification; (c) data that were
misleading; and (d) data that were falsely reported to us.
To comply with your request at that hearing, I am providing
herewith documents and information on two issues: the data
from which we were excluded concerning the reliability of the
Peacekeeper missile warhead, and the data which we found to
be false concerning the radar cross section (RCS) of the B-1B
bomber.
Denial of Data on Peacekeeper
Warhead Reliability
GAO first requested up-to-date data on Peacekeeper warhead
reliability in June 1989, which had been reported
periodically in reports known as "CINCEVALS." We had always
had access to these CINCEVAL reports in the past and were not
given a clear understanding of why we couldn't see this one.
Its classification level was one to which we normally have
access.
We repeated our request to various DOD offices, moving upward
in the chain of command, over a period of a year. At each
level, we were refused the portions of the report on the
estimate of Peacekeeper warhead reliability. As you can
see in enclosure I, General Chain, then commander of the
Strategic Air Command, objected to release of the data on the
65
grounds that it would "pose a grave threat to national
security." Although GAO did then receive a copy of the
September 1989 Peacekeeper CINCEVAL (the report at issue) ,
the section on warhead reliability was blacked PUt.
On August 3, 1990, GAO sent its first data demand letter to
the Secretary of Defense for the data in question (see
enclosure II). This was denied by DOD in a letter signed by
its General Counsel, on August 31, 1990 (enclosure III). GAO
responded with a second demand letter on October 26 to
Secretary of Defense Cheney and to OMB Director Daman
(enclosure IV) . This request was again subsequently denied
in a letter signed by Director Darman (enclosure V) . After
this last refusal, we ceased our efforts to enable GAO
evaluators to review these data and we have noted in our
report our inability to corroborate the high estimates of
both Peacekeeper's warhead reliability, and hence the
missile's overall weapon system reliability.
B-1B Radar Cross Section fRCS)
Enclosure VI, the memorandum I referred to in my testimony
last Thursday, is one of two sources for our realization that
false data were being provided to us and accurate data
withheld. This memorandum was found by GAO evaluators in a
package of written DOD comments that responded to a 1989
draft GAO report. I draw your attention first, to the data
given in paragraph l.b.(l) on the B-lB's RCS versus the data
given directly below under l.b. (1) "CONCERN." which are
accurate, but not to be released to us; and second, to the
data given under l.b. (2) versus the accurate data given
directly below under l.b. (2), "CONCERN." The language given
under the two CONCERN headings specified the accurate RCS
measurement of an actual B-1B and provided an interpretation
of these accurate data.
The data in the two passages marked CONCERN — that is, the
information "not intended for release to the GAO" — were
nonetheless identified and confirmed through independent
inquiry and analysis by GAO evaluators, and were incorporated
in a draft of our report, U.S. Strategic Triad: — Modernizing
Strategic Bombers and Their Missiles. This draft report was
then transmitted to the Department of Defense for its
official comments on June 11, 1991, and was the subject of
several meetings of DOD and GAO representatives. These
meetings were held to provide DOD an opportunity to supply
official oral comments in response to the draft and
documentation to support those comments. One of those
meetings occurred on July 24, 1991 (the attendees to that
meeting are shown in enclosure VII). At that meeting, the
project manager for our draft report was given draft written
DOD comments, which included the document at enclosure VIII.
66
This 1991 document is the second source I mentioned earlier:
it shows that 2 years later, incorrect b-ib Res data were
•till being offered to GAO. In addition, over the course of
DOD's official oral comments we were criticized for the
accurate data we used in our report and told to revise our
draft to conform to the inaccurate data shown in enclosure
VIII.
During these discussions, no new data or documentation were
provided to us that could have refuted our RCS calculations
and finding. Moreover, a DOD representative at the official
oral comments session, stated that he believed both our data
and our interpretation of them to be accurate. Still
further, in final written comments to our draft report, the
Department of Defense recognized the accuracy of our finding,
based on the RCS measurements we reported, and stated in its
comments, "the DOD agrees with the GAO" concerning our
comparison of the B-52 and the B-IB on the issue of radar
cross section.
Thank you again for your careful review of our study and for
the important hearing which resulted from it. If you should
have further questions or comments about the information we
have provided, please call me at 512-2900 or Kwai-Cheung
Chan, Director for Program Evaluation in Physical Systems
Areas at 512-3092.
Sincerely yours,
Eleanor Chelimsky
Assistant Comptroller General
Enclosures
67
II
ENCLOSURE I
r.rti
.1 : !
v.
ENCLOSURE I
DEPARTMENT OF THE AIR FORCE
MIAOOU*RTCRS STRATEGIC AIR CCMMAnO
OFfUTT AIR FORCE BASE. NEBRASKA 681 13
Memo for Record
Subject: GAD Request for Top Secret Docuroe-rt
1. I have reviewed the Corcptroller General's letter to Secretary Cheney
(B-240665, 3 Aug 90) . After careful consideration of the document in
question, "Evaluation of Weapon System Capabilities - PEACEKEEPER in hONUTEMAN
Silos (U) ," 18 Sep 89, I object to the release of this document to the GAO.
2. Information contained in this report is extremely sensitive, limited
access, TOP SBCRET/Restricted Data, and must be carefully controlled. Release
of this information would pose a grave threat to national security.
JOHN T. CHAIN, JR.
General, USAF
Connander in Chief
68
ENCLOSURE 11
ENCLOSURE 1}
Comptroller General
of the United State*
Waahlaftoa. D.C WMa
vAwLtV.
f/L*jrsd
B-240665
August 3, 1990
The Honorable Richard B. Cheney
The Secretary of Defense
Dear Mr. Secretary:
Pursuant to 31 U.S.C. 716(b), I am writing to request
immediate access to the most recent complete and unedited
(TS/RD) Peacekeeper CINCEVAL. The title of this report is
"Evaluation of Weapon System Capabilities, Peacekeeper in
Minuteraan Silos," dated 18 September 1989 (ICBM Evaluation
Branch, 549 WSES, SAC). In order to conduct our evaluation
of strategic nuclear weapons (973266), we need to have
access to the latest CINCEVAL of the Peacekeeper missile.
I
As you know, we were previously provided with an extensively
[edited version of this report. But because all information
'regarding the actual current reliability of the weapon
system was deleted, it does not meet our needs. Without
this information, we cannot provide the Congress with ar.
'informed analysis of the true reliability of the weapon
'system.
We have previously been given
'unedited versions of all other
nuclear weapons which we have
this project, including the pr
document we now seek is not a
document. GAO staff who will
hold the appropriate security
course safeguard it in accorda
requirements. " CINCSAC and the
bof.b - jdicated that they have
this document to GAO.
access to the complete
CINCEVALS for strategic
requested in connection ■ •_ -r,
ior Peacekeeper CINCEVAL. Th-
planning nor a war-fighting
have access to the document
clearances and we will of
nee with established
Department of Energy have
no objection to the release o;
This review is being conducted pursuant to GAO's author:. t«
under 31 U.S.C. 712 and 31 U.S.C. 717. We have a right :f
'access to the information requested under 31 U.S.C. 716: ' .
Pursuant to 31 D.S.C. 716, your response to this letter is
due within 20 days. If full access to the requested
information is not granted, you are required to furnish a
description of any information withheld, and the reasons for
withholding it.
Sincerely yours,
^ t
— i '
Charles A. Bowsher
Comptroller General
of the United States
69
ENCLOSURE III w.fcUfci .. . .1
■SECRET
ENCLOSURE III
GENERAL COUNSEL OF THE DEPARTMENT OF DEFENSE
WASHINGTON. D C 20301-KOO
31 August 1990
Honorable Charles A. Bowsher
Comptroller General
441 G Street, N.W.
Washington, DC 20548
Dear Mr. Bowsher:
(U) This Is In response to your letter dated August 3, 1990, which was
received by the Office of the Secretary of Defense on August 13, 1990 (GAO File B-
240665).
(U) First, It appears that your letter was based on a serious
misunderstanding concerning the positions of the Commander In Chief of the
Strategic Air Command and the Department of Energy. Your letter represents that
"CINCSAC and the Department of Energy have both Indicated that they have no
objection to the release of this document ["Evaluation of Weapon System
Capabilities, Peacekeeper In Mlnuteman Silos," dated 9/18/89] to GAO." This Is not
correct. CINCSAC objects to the release of this document to GAO (Tab A), and the
Department of Energy can find no evidence that anyone In the Department made any
such representation to GAO. In fact, the Department of Energy defers to the views
of the Department of Defense on the question of the release of the report to GAO
(Tab B).
i
(U) Third, as we discussed last December, the Secretary personally briefed
key Congressional leaders on this matter.
(U) If you feel that It would be helpful, I am available to meet with you and
your staff to discuss this Issue further and to reiterate the Important national
security concerns and factual Issues underlying the Department's position.
Sincerely.
Terrence O'Donneli
70
ENCLOSURE IV ENCLOSURE IV
B-240665.2
October 26, 1990
The Honorable Dick Cheney
The Secretary of Defense
Dear Mr. Secretary:
Pursuant to the last sentence of section 716(b)(1) of Title
31, United States Code, I am submitting this report on your
failure to provide access to certain records to the General
Accounting Office (GAO) .
Section 716(b) of Title 31, United States Code, establishes
mechanisms for the resolution of GAO access to records
problems involving federal agencies. Subsection 716(b)(1)
provides that when access to any records of a federal agency
is not made available to GAO within a reasonable time, the
Comptroller General may make a written request for such
records to the agency head. If full access is not afforded
within 20 days from receipt of that request, "the Comptroller
General may file a report with the President, the Director of
the Office of Management and Budget, the Attorney General, tht
head of the agency, and Congress."
After 20 days following the filing of this report, the
Comptroller General may apply for judicial enforcement of the
access request, subject to the right of the President or the
Director of the Office of Management and Budget to preclude an
enforcement action for certain kinds of records.
GAO staff had asked Department of Defense officials for access
to the most recent complete and unedited (TS/RD) Peacekeeper
CINCEVAL, entitled "Evaluation of Weapon System Capabilities,
Peacekeeper in Minuteman Silos," dated 18 Sept. 1989 CC3M
Evaluation Branch, 549 WSES, SAC) .
Although several requests were made, GAO was net grar.ted
access to the information. However, in conversations over a
number of months we were advised by ycur General Counsel that
the report would be made available upon completion of certai.i
sensitive requirements. The delay proved to be unreasonably
long in the context of our need for the information involved,
and 1, therefore, on August 3, 1990, made a written request
for the documents sursuant to 31 U.S.C. § 716(b) (Enclosure
I) .
71
•I •
We war* not afforded full access to the requested racords
within 20 days following my writtan request. On August 31,
1990, tha Ganaral Counsal of the Department of Defense replied
to sty request in a classified letter. I have reviewed the
representations in that letter. I remain persuaded that GAO
needs tha information and is justified in invoking its legal
authority to compel disclosure.
Since GAO has a legal right of access to the requested
documents, and since full access to these records was not
afforded within 20 days following our August 3 request, 1 now
submit my report of the matter to you and tha other designated
officials, pursuant to 31 U.S.C. 5 716(b)(1).
Sincerely yours
Comptroller General
of the United States
Enclosure
72
' ^
B-240665.2
October 26, 1990
The Honorable Richard G. Darman
Director, Office of Management
and Budget
Oear Mr. Darman:
Pursuant to the last sentence of section 716(b)(1) of Title
31, United States Code, I an submitting this report on the
failure of the Secretary of Defense to provide access to
certain records to the General Accounting Office (GAO) .
Section 716(b) of Title 31, United States Code, establishes
mechanisms for the resolution of GAO access to records
problems involving federal agencies. Subsection 716(b)(1)
provides that when access to any records of a federal agency
is not made available to GAO within a reasonable time, the
Comptroller General may make a written request for such
records to the agency head. If full access is not afforded
within 20 days from receipt of that request, "the Comptroller
General may file a report with the President, the Director of
the Office of Management and Budget, the Attorney General, the
head of the agency, and Congress."
After 20 days following the filing of this report, the
Comptroller General may apply for judicial enforcement of the
access request, subject to the right of the President or the
Director of the Office of Management and 3udget to preclude an
enforcement action for certain kinds of records .
GAO staff had asked Department of Defense officials for access
to the most recent complete and unedited (TS/R2) Peacekeeper
CINCEVA1, entitled "Evaluation of Weapon System Capabilities,
Peacekeeper in Minuteman Silos," dated IB Sept. 19B9 CC3M
Evaluation Branch, 549 WSES, SAC) .
Although several requests were made, GAO was r.ot granted
access to the information. However, in conversations over a
number of months we were advised by the General Counsel of the
Department of Qefense that the report would be made available
upon completion of certain sensitive requirements. The delay
proved to be unreasonably long in the context of our need for
the information involved, and I, therefore, on August 3, 1 59C ■
made a written request for the documents pursuant to 31 U.5.C.
S 716(b) (Enclosure I) .
73
M« ••re not afforded full access to the requested records
within 20 days following my written request. On August 31,
1990, th« General Counsel of the Department of Defense replied
to my request in a classified letter. I have reviewed the
representations in that letter. I remain persuaded that GAO
needs the information and is Justified in invoicing its legal
authority to compel disclosure.
Since GAO has a legal right of access to the requerted
documents, and since full access to these records was not
afforded within 20 days following our August 3 requtat, I no--
submit my report of the matter to you and the other designated
officials, pursuant to 31 U.3.C. S 716(b)(1).
Sincerely yours,
Comptroller General
of the United States
Enclosure
74
ENCLOSURE V ENCLOSURE V
EXECUTIVE OFFICE OF THE PRESIDENT
OFFICE OF MANAGEMENT AND BUDGET
WASHINGTON. C.C. 20503
the director November 23, 1990
Honorable Charles A. Bowsher
Comptroller General of the United States
U.S. General Accounting Office
Washington, D.C. 20548
Re: B-240665.2
Dear Mr. Bowsher:
This letter constitutes a certification, pursuant to 31
U.S.C. Section 716(d), that the redacted portions of the
document referred to in your report dated October 26, 1990,
which was received by the White House on November 9, 1990,
could be withheld under 5 U.S.C. Section 552(b)(5), and th*t
disclosure reasonably could be expected to impair
substantially the operations of the Government.
The document referred to in your report entitled
"Evaluation of Weapon System Capabilities — Peacekeeper in
Minuteman Silos," dated September 18, 1989, is a Restricted,
Top Secret, uniquely sensitive Strategic Air Command report.
A redacted copy of the report was provided to the General
Accounting Office (GAO) . The redacted portions of the report
are highly classified and contain estimates on the W87
nuclear warhead.
5 U.S.C. Section 552(b)(5) authorizes the withholding of
all "inter-agency or intra-agency memorandums or letters
which would not be available by law to a party other than an
agency in litigation with the agency." Thus, Exemption 5
exempts from mandatory disclosure all documents that are
"normally privileged in the civil discovery context." NLRB
v. Sears. Roebuck & Co. . 421 U.S. 132, 149 (1975). See
United States v. Weber Aircraft Corp., 465 U.S. 792, 798-804
(1984) (all privileges that are well recognized in the case
law are incorporated by Exemption 5) .
As described above, the redacted portions of the
Strategic Air Command report constitute a staff evaluation
for the Secretary of Defense and other Defense Department
decisionmakers of the capabilities of a very sensitive
weapons system for use as planning factors in developing and
updating the Single Integrated Operations Plan (SIOP) . The
SIOP is among the most sensitive of the nation's military
secrets. The redacted portions thus may be withheld under
Exemption 5 covering predecisional, deliberative mercrranda
75
and reports for agency decisionmakers. See Sears. Roebuck.
421 U.S. at 150-151. See also United States
v. Reynolds. 345 U.S. 1, 10 (1953) (privilege covering
"military matters, which, in the interest of national
security, should not be divulged.").
The Secretary of Defense has determined that disclosure
of the redacted portions of this document to the GAO
reasonably could be expected to impair substantially the
operations of the Government. The redacted information is
deliberative material used to make weapons allocation and
targeting determinations by DOD authorities charged with tht.
oversight of the SIOP. The redacted material consists of
uniquely sensitive estimates to assist in the management of
the nation's nuclear deterrent aimed at holding Soviet
targets at risk. Reporting to Congress on this extremely
sensitive information appropriately has been limited to a
briefing by the Secretary of Defense of the leadership of .. •
House and Senate Armed Services Committees.
• I agree with the Secretary's determination and hereby
make the certification required by 31 U.S.C. Section 716(d)
Since all certifications under this section must be provitft'.
to Congress, as well as the Comptroller General, copies o:_
this letter are being submitted to the President of the
Senate and the Speaker of the House of Representatives.
With best regards,
Richard G. Darman
76
ENCLOSURE VI
• ufcJtc:
DEPARTMENT CFTWE AIR ":RCi
MEAOOUAR-TERS STRATEGIC AIR COMMANO
0«TUTT AIR FORCE BASE. NEBRASKA • •113-ftOOl
ENCLOSURE VI
Is 2 8 J'JN 1cga
GAO Draft Report, on "E-1B Mission Effectiveness" (U)
AC
1. (Uj We "nave reviewed the GAO ' s craft report, on "3-iB Mission
Effectiveness" and coordinate with the following comments:
a. ( )
b. (U) The following two comments were previously stated in
INA's 18 Mar 89 letter to XOB for their response to the earlier
draft. They are included here for the record and reflect the
standing DCS/lntelligence position: however, they are not intended
for release to the GAO.
(1) ( )
( ) CONCERN :
(2) ( )
( ) CONCERN :
.r.rr Ms -. Lsrrv
2. (U) If you have further questions, you may contact Maj
Thomson, IN'AD, 42476.
/
//t
-'AMES P. ROOT. Ccione!, USAr
Assi DCS-lnieilisancs
ClASSiriEr EY: E-13 Security C-u-ce
m mm
77
j
\
•vN
r»
.
^
•
'
:•.>:
1
i
c
0
i
p
u
p.
Is
'
tM —
UI ^r
1 ^~
M
Si
«0
t
\3
i
r
••
<
XT
r
>
N
^■^
\5
4
'
-^w
u.
3
^
ft.
J \
3
s
CO.
M
u
V,
5
7-
^
4
"2
h
>
O
-
3
u
o-
{
\5
10
6,
'•U
It
J
s
^
a
-J
6
u.
P
V
P»
_y
l T
Q
s
5
V)
^
3
o-
r<»
^
if
3~
>
m
■ >
h
<
^5
H
Z
U
z
o
H
14
M
5
3
ae
o
^
^
VI
n.
g
o
r=:
ATTENDEES
TWEEN CAO HEPAE:
ONENTS I CAO
■^5
<3
C
15
u.
<
c
i
v/
h
H
s 5
? *
■=k
^1
S3
->
^0
•
z
d
*
<
^
sjk.
■
£ a
Vff
<>
•\
J
-> y
O
»>
«
1.
^*
UJ 3
Si
3
u u
v.
fel/)
J"
k
o
a
t
i
H
V\
-3
5
V5
i
4
s
•4.
•->
^
, i-
^ *
Hi >
c
<
X
«
•
s
^
^v
5
3
7
. r
"<3
8
«0
t'
1
J
■j
>
V
V
5
^5
•
' 1-
A
I
o
■j
0
00
VJ
i
>-
1
'J
V
II
V
tu
I
Q§
<9
■c
o
-j
j
c
^5
**
d
"*^N
1
i
% !
g »
78
to
2
O
CJ
UJ
CO
CO
to
c
C£
CJ
<
Q
<
-J
<
o
>
UJ
lii
<
LO
CO
CQ
CO
cc
o
z
Q
o
z
UJ
o
2
CJ
a.
co
Z>
O
o
-J
Q
UJ
O
UJ
D
—i
CO
CC
UJ
o
UJ
UJ
CQ
>
UJ
Z
O
x
=*&*=»
3
SI
O
J
J
2
-3
79
TAB
NEWS
B-lBandtheGAO, JUNE IS, 1593
Allegations that the Air Force deliberately provided false informa-
tion to the GAO on the B-1B is not true. The GAO has been in
possession of aD B-1B Radar Cross Section data since July 1989. in
an August 1989 meeting between the Air Farce, DOD and the
GAO the topic of Radar Cross Section data was addressed. The
allegation of withholding information was discussed and the Air
Force officer who attended the meeting left believing the GAO
knew that the Air Force did not intend to deceive the GAO
regarding the release of any RCS data.
TALKING POINTS
- There was a June 24, 1989 letter, classified secret, which discussed
RCS and did state the figures were not intended for release to
GAO. The letter was the opinion of one action officer and did not
necessarily reflect the views of the Air Force or DOD.
- A July 19, 1989 letter from the GAO, to JCS* Strategic Operations
Division, also classified secret, contains the figures cited in the 24
June letter. The July 19 letter states the GAO received complete
and accurate information during their process of reviewing DOD's
comments on the GAO's draft report on the B-l.
- Finally, in an August 10, 1989 unclassified memo from SAF/AQ
(Acquisition) to Air Force XOOT (operations) the 8 August 89
meeting between the Air Force, DOD and GAO is summarized.
The Air Force officer who attended and made notes at the meeting
said me allegations about RCS data were discussed openly and
believed that, "More imporfntly, the GAO left the meeting
knowing mat the Air Force did not intend to deceive the GAO
regarding the release of any RCS data."
Source: SAF/AQ
AX> Maj Tom LaRock
80
5: 2 i
^ •- J!
5 ° i
°
c c -z.
: -• c o
a s !j x o c
is 3
I 5 3
U
^ I
— X 2
O -
c — . S > ; O .=
• r -s ~ o o ■ v ■
=0
S3
<
_
os-i-
~ c 3— a -S
<
a
m «• 3 a
2 zo ro--=
^—^
<s .. < =
CO
CJ
— "i.^-c
5?" t^ C> — - « 5
co
c
» z ~
K .=
X ~ =.
CO
5— Si ?
3 ~ ~ *°
2- — ~
5^-co
Se "
— r — ^j — j _j ; - t-L3
S S3!
= =0
2 — Z H~S»?.i
u_^ 2 = 5 — ™ = jj <
"1 lI""-2^ i-2
O
~
— /"N
_ as
-<
e Tt
>>
VI
50
3 2
C
—
V —
_,
— 0
. - £5
n
CO
— as
5<
0
V
_>>
r
■5 2
-
— _r;
c
5* —
^ — >
2 c
ci
•J
*o
*~
"*"
- CJ
—
a
0
4* —
X
u
w
a "=
C
ci u
0
-j
5 ?
— -•
M
~~.
G-£
y>
^\ a
-*
< "
~
^ —
— ■
en
U =
?
C S
3
«s r
_
- "=
'-)
" 5
— CJ
=■
Mil
81
UNCLASSIFIED
APPENDIX II APPENDIX II
COMMENTS FROM THE DIRECTOR OF DEFENSE
RESEARCH AND ENGINEERTW,
^»TJ*L
DIRECTOR OF DCFCNSElMSSANCH AND CNOINCERINS
WASHIN6TON.eC 10101-3010
1 0 AUG 1989
Mr. rr«nk C. Conahan
Assistant Comptroller General
National Security and
International Affairs Sivuion
0. S. General Accounting Office
Washington, DC 20348
Dear Mr. Conahan:
»i» is the Dcpartaene of Defense (OoO) response to the
BOMBERS: .ssues Related to a-la as a Penetrating Sooner " d«-.rf
JOM I, 1988 (CAO Cod. 393«33), OSD Case 1917. «SoS.ttal5nf
generally concurs with the draft report? oepartnent
(-n„J?'.D°P*rlaent.fu? «"><Sueted a thorough review of the
UA!?« Je?t"fd,ntiei*S c«"eiencit« of tzi UO-ISXA systea. -h.
ALO-ISU is intended to coapl.cu.nt the inherent B-ls fliSSt
performance characteristics that contribute to the penetration
been perforaed that support the Dep.rt.enfs plan to cor *<'
deficiencies and aaxisise the aircrew's abilitv to detee-
are SSfd^M^TI "fT *"»« ««ce.aes . near-tern DoD efforts
ace toward neetmg baseline defensive system reouireaents
!f:! :;^! "•"""nee. Suture enhancements miqht be rem i red to
aeet teennology edvaneeaer.ta or clanging threats Ae the -enort
ffn3^"- <"" SP**5"<= enhanceme«rwill".5;nd^ JSTeSSS?
roles that say oe assignee to the e-1*. suture
on July is. i-es. the GAO asaed tae Of Sice of ' the join.
Chief, of Staff and the Strategic Air^oanind to -.««« i-ib
cross sect. on differed froa design requlreaents. The
82
tin- -• 4 ■»«••«-• _ri
APPENDIX II APPENDIX II
referenced "new information" is actual flight-test-derived radar
cross section values, which in Tact had previously been provided
to the CAO. The flight test data did indicate a somewhat higher
radar cross section than that measured in model tests (which
were the basis for the contractor meeting specifications). The
flight test data also showed, as expected, significant
variations with aspect angle and frequency. However, the
modestly higher radar cross section at critical sectors and
frequencies does not significantly affect the penetration
effectiveness of tn« B-IB. -The effectiveness of the 3-lB as a
penetrating bomber derives from a combination of its low radar
cross section, high penetration speed, and low altitude
capability. Because of its low altitude, the point at which
ground-based radars, both those for surface-to-air missiles and
those vectoring interceptor aircraft, will detect the S-1B are a
result of line-of-sight detection from breaking the horizon at
close ranges and are not influenced by its radar cross section.
Consequently, the "new information" on tbe aircraft's radar
cross section should not alter the findings of the current CAO
report.
Specific DoO comments on the draft report findings are
provided in trie enclosure. The DoD appreciates the opportunity
to comment en the draft report.
Sincerclv
Robert"c. Otincari
Enclosure
34
83
finftd gates Senate
WASHIN8T0N. DC 208 10
June 10/ 1992
Dear Colleague:
We are sending this letter to the members of the Senate
Armed Services Committee and the Defense Appropriations
Subcommittee as substantive information on the capabilities of
the 3-le Lancer bomber as assessed by General John M. Loh, USAF,
Commander Air Combat Command, and General George L. Butler,
Commander-in-Chief, strategic Command, the two senior commanders
who have responsibility for employing the aircraft. General Loh
and General Butler were responding to our requests for their
evaluations of the B-l's conventional capabilities and of its
operational utility in the heavy bomber force.
In our view the
ijflMikM£iM**fted resistance in tbe Congress to meaningful
modifications that will allow best employment of its inherent
capabilities. General Loh and General Butler agree, and in
their letters they strongly urge enhancement of the B-l
conventional capability. Among other needs to qualify this
superb machine for best employment is integration of 1760
wiring, inclusion of a Global Positioning Satellite receiver,
and qualification for inertially-aided munitions.
Importantly, during the week of June 15 the Air Force will
deliver to the Congress the Bomber Roadmap in which General Loh
and General Butler played substantial roles. The Roadmap will
address the integration of the 3-1, B-2, and B-52 bombers in a
comprehensive assessment of U.S. heavy bomber operational
capabilities, and reportedly will reflect General Loh and
General Butler's strong endorsement of 3-1 conventional mission
needs.
Our letters to General Loh and General Butler and their
replies are attached for information; a classified (secret)
supplement to General Loh's letter comparing the weapons and
range capabilities of the B-l and the B-52H is also available to
you and your staff in S-407.
In sunroary, we uxge you to fully support funding of B-l
upgrades in the FY-93 Defense Authorization and Appropriations
bills as an outstanding investment in much needed additional
qualifications for this already proven aircraft.
We welcome discussion on this matter either with us or our
staffs.
Best regards.
Sincerely,
Alan J.
United
Senator
Richard C. Shelbj
United States Senator
84
DEPARTMENT OF THE AIR FORCE
HEADQUARTER! AIR COMBAT COMMAND
LANOUY AIR FORCE BASE, VIR3INIA
JUK 5 t932
OFFICE OF THE COMMANDER
Langley AFB VA 23665-5555
Honorable John Glenn
United States Senate
Washington DC 20510-0001
Dear Senator Glenn
Thank you lor your letter of 22 May 1992. The B-l questions you
raised are timely, and I welcome the opportunity to clarify
these natters.
As the Gulf War proved, the demise of the Soviet union did not
eliminate all threats to American interes-s. On the contrary,
the world iB less stable now than during the Cold War and with
the proliferation of sophisticated weapons around the world, the
threats we face are varied, complex, and plentiful, with fewer
forces overseas and knowing that our allies' interests may
converge with our own, our ability to respond immediately and
forcefully in future crises is imperative.
This demanding set of criteria explains our emphasis on the
bomber's conventional capabilities. Only the bomber force can
provide immediate, massed firepower in the opening phase of a
conflict anywhere on the globe from home or from a theater
location. With the B-52 force attacking from primarily a
standoff position, and a planned force of only 20 B-2s, it is
imperative the conventional capability of the B-l be enhanced.
With 96 in the force today, it will be the workhorse of our
conventional heavy bomber force. Although the B-l was
originally optimized for nuclear operations, it was designed for
utilization in both the nuclear and conventional role and,
therefore, has tremendous inherent capabilities for conventional
operations which can be fully exploited. The b-1'b combination
of very high speed, long range, high payload and low altitude
terrain following avionics makes it ideally suited for
conventional missions, with modifications. These modifications
will only improve the B-l's total war-fighting capability. They
will not diminish the ability to perform its nuclear mission
which can be fully retained.
Today, the B-l's conventional capability is limited to carrying
feneral purpose bombs; however, we intend to
to carr/gHBHHQ|M^i£icn munitions. By
comparison, an F-lll—our larges^^^^Wer platform for these
weapons— carries only OTMHHMl^VMPtMMlMP~~
85
I'm sending a classified chart separately to illustrate better
why I believe it is prudent to modify this aircraft to reach its
full capabilities. The chart compares the B-52H to the B-l,
showing the weapons and range capabilities of the two aircraft.
In some cases, you will notice the aircraft are very similar;
however, there are dramatic differences in the two aircraft's
war-fighting capabilities.
First, the B-l is much more compatible with composite force
^operations. Its speed and radar signature are similar to the
other aircraft in a strike force package, unlike the much older
■ggpMHMQSffKThe B-l can
also operate "from the same type of airfields as our fighter
aircraft, using narrower taxiways and shorter runways. During
speserr, storm, the B-l could have operated from more than 30
airfields within the theater, while the B-52 was restricted to
less than 10.
Secondly, the B-l, even with its current electronic counter-
measures (ECM) limitations, is a more capable and survivable
platform than the B-52. If we are able to proceed with the
planned B-l ECM program, we will be able to exploit its inherent
advantages of high speed and lower radar signature better, with
the ECM, the B-l can penetrate defenses and strike targets
unescorted using direct attack munitions, while the B-52 cannot.
The B-52 could only be used to directly overfly targets in areas
where the air defenses have been severely degraded.
Consequently, we will use the B-52 primarily in a standoff role
necessitating the use of significantly more expensive missiles
until defenses have been eliminated when it can add mass and
reach to follow on sustained operations.
The
we want to make to the B-l ECM suite affects.
Irst, we want all the aircraft to have
■Pwe plan to improve the ^rews '
situational awareness. This will allow them to
CHMg|H00HBHHBI^VHHHBV'v€:ntior>al threats, while
simultaneously detecting modern radars both Russian and non-
Fussian. Finally, we will iif__
thereby, increasing employment options ana survivability against
the threats we expect it to face.
Third, the B-l has many unique capabilities the B-52 can never
match. It has the speed advantage cf_routinely attacking
targets at «MWSBBBHHHB(IBHWWHWWBli*BWBW|«P«»-e suiting
in much longer threat exposure. Also, the B-l can use its
automatic terrain following system to ingress to its targets at
2 00 feet. These features, along with its greater bomb load,
better accuracy, and high resolution radar, give the B-l
enormous potential to provide significant power projection well
into the next century.
86
4
I
Finally, it is important to note the average
VMMMMMtfHMl It will cost us less to maintain a
B-l well into the twenty-first century than to continue to shore
up a aid-twentieth century design. To date, restrictions which
have denied us the baseline program funds to provide for spare
parts and establish organic repair capability have kept B-l
aircraft availability rates artificially low. These low rates
&n<3Jth£_JpBMMriesing|tigggaA^ftptf^£iBeve given the B-l an
leserved negative Characterization.. .The B-l . .availability
■*Tates xor its_first 6 years of operations are similar to other
^sonpieoraTrcraft at"this~ stage of maturity. Therefore, we are
convinced adequate B-l provisioning would increase aircraft
availability rates comparable to the B-52 and our current
fighters.
Delaying funds for our proposed B-l conventional enhancements
program, will leave the B-l with its current MK-82 soft, area
target kill capability. As you are aware, this program
includes: MIL-STD 1760 avionics architecture, ECM upgrades,
full integration of the GPS system into the offensive avionics,
qualification of the Joint Direct Attack Munitions, improved
radios, and an upgraded computer system. Additionally, since
the proposed ECM improvements are not currently in' production, '
.any funding- delay will*. iTSg the start- date,- affecting the other
/improvements in capability: Delays in ECM funding also- leave us"
' with 13 aircraft" with ho^defensive avionics at all, and we will
be unable to decrease the roan-hours required to trouble-shoot
system problems.
Hopefully, this information, will answer your concerns. With
carefully planned modifications, the B-l will add to our ability
to project conventional military power significantly and well
into the future.
/
Sincerely
JOIK^M. LOH
General, USAF
Commander
87
DEFENSE WEEK Monday, Octobtr .'(5. 1992
The All Force over the
Uut decade has incorrectly
stated the stealthy, head-
on profile of the B-l
bomber, claiming It would
appear smaller oa enemy
radar screens than the
data Indicates, according
to the unclassified
_ tiont of a new secret stud;
According to
cross section
data and pilofn
B-l Not As Stealthy
As Air Force Claims,
Says GAO Review
BY TONY CAPACCIO
"We use the front-view
radar cross section, which
we have done consisten-
tly," Camana said. "If
they take issue with that,
it's a problem with their
interpretation of the data
the Air Force is using be-
cause we have been using
head-on radar
Force offi-
congressional testimony
and interviews that the B-l, with its fully-swept back
76-foot wing span, has a profile about the size of an F- 1 6
fighter, with its 32-foot wing span.
While never justified exclusively on the basis of its
stealthy characteristics, the B-l was billed as a marked im-
provement over the B-52 in terms of radar cross section.
But the cross section issue strikes at the heart of questions
about the B-l 's survivability in low-level nuclear bomb-
ing runs over the once-Soviet Union.
Until the Soviet empire crumbled, the jury was still out
on whether the B-l could meet its low-level operations
requirements.
"GAO found that the Air Force has incorrectly repor-
ted the radar cross section of the B-l B. ..Although the Air
Force states the B-l's head-on radar cross section is no
greater than (deleted RCS figure), GAO found that an ac-
tual head-on measurement was [deleted RCS figure], de-
pending on the type of radar performing the detection,"
GAO wrote.
Based on their analysis of classified stealth data, GAO
investigators concluded the B-l's RCS was significantly
larger than service claims, said a congressional source
familiar with the study.
The unclassified GAO conclusions are contained in a
classified study entitled, "U.S. Strategic Triad: Moder-
nizing Strategic Bombers and Their Missiles," It was
provided September 28 to Rep. Dante Fascell (7>Fla.),
the retiring chairman of the House F oreign Affairs
Committee, as part of an eight-volume study on the nu-
clear triad.
Fascell placed as unclassified summary of the entire
study into the Congressional Record, but the B-l stealth
discussion has not been made public.
News of the B-l RCS controversy is the latest in a
string of unflattering revelations to dog the B-l. But de-
spite major funding cuts suggested by the Senate Armed
Services Committee and its appropriations counterpart,
congressional defense conferees decided to approve
$304. 1 million of the Pentagon's S35S .9 million fiscal
1993 procurement and rtsearcoAJeveiopment request.
And Air Force officials at a B-l ' 'summit' ' meeting
late last month tentatively approved an approximately
SI .9 biJlioa. three-part conventional B- 1 upgrade pro-
gram that fleshes out the details first unveiled in June in
the service's "Bomber Roadmap."
"We really don't need to get into numbers" of B-l
RCS measurement, said Ma). Gen. Patrick Camana, Air
Force director of strategic acqmtjtipj
Carotins, had direct involvement with GAO during
compilation of the bomber- report-
that data all along. ..I guess we would have a basic dis-
agreement on this issue," Caruana said.
According to GAO, even if the B-I 's head-on RCS was
as small as the Air Force claimed, its apparently larger
side and rear profiles negate any advantage against So-
viet-style radar provided by the more compact frontal
view.
The purportedly small head-on view "can be a mislead-
ing description, since from the side and rear GAO found
that the B-l's RCS is [figure deleted]. Clearly, this makes
the smaller head-on measurement much less meaningful
since it is unlikely that air-to-air and air-to-ground eng-
agements will only occur with the aircraft flying head-on
to Soviet radars," the report said.
Concerning the recently completed B-l "summit,"
Caruana said the Air Force plans to begin lethality en-
hancements this fiscal year. The enhancements call for
installation of Global Positioning System equipment for
navigation and weapons Improvement and Military Stan-
dard 1760 wiring that allows for the carriage of conven-
tional ordnance such as the Joint Direct Attack Munition
and Tri-Service Attack Munition.
According to Caruana, the summit participants agreed
to press forward with improved countermeasures and
"situational awareness" gear. He declined to explain
whether the Air Force will abandon the so-called CORE
electronic countermeasures (ECM) system— a system that
despite years of development has yet to meet all of its
operational and maintenance requirements.
"You will see that we stayed very close to the
[bomber] roadmap. But what we filled in, we will be
more specific in timing the phasing of lethality and
ECM improvements," he said.
"W« are not taking a 'dean sheet* review but based on
combat requirements from the Air Combat Command
look at what system will give the best situational aware-
ness," Caruana said.
"The roadmap addressed the first five days of a cam-
paign and a sustained operation. The summit refined that
even more. We will see a better defined statement for the
lethality, survivability and provisioning than you did In
the roadmap, " be said.
Congressional approval of virtually all the Pentagon's
funding request m the face of B-l criticism "suggests that
they are ready to look at this airplane as they look at
every other airplane," Caruana said, instead of as the
much-maligned bomber of the Reagan years.
Caruana has said Congress is unwilling to let the B-l
"out of the penalty box," but with this year's funding
may release it.
"1 think the airplane is moving out of that [penalty
box] cawgory," he said. "Given the $20 billion [in 19E2
dollars] spent on the program, it's to everyone's advan-
tage for us to press down this very calculated path with a
same plan to continue to enhance the bomber's capabi-
lity "
88
mtmam «. m
Bmtd States Senate
COMMITTEE ON
GOVERNMENTAL MMM
WASHINGTON. DC 20f 1O42S0
June 28, 1993
The Honorable Les Aspin
Secretary of Defense
Washington, D.C. 20301
Dear Mr. Secretary:
On June 10, 1993, I chaired a hearing of the Governmental
Affairs Committee on the General Accounting Office (GAO) Evaluation
of the O.S. Strategic Modernization Program.
At that hearing, I was shocked to learn, in response to a
question I asked of Assistant Comptroller General Eleanor
Chelimsky, that the Department of Defense (DOD) had provided
falsified data to the GAO concerning the radar cross section (RCS)
of the B-1B bomber.
This deception was discovered by the GAO when the Strategic
Air Command inadvertently delivered a memo blatantly describing the
deception (and showing the correct data) as part of a package of
written DOD comments responding to a 1989 draft, of a GAO report on
the B-1B. Subsequent meetings between the GAO and the Air Porce
resulted in changes to the report.
In 1991, GAO was deeply into its study of triad systems and
received draft written comments from the Air Porce stating in
substance that GAO's measurement data and interpretation thereof
on the B-1B was wrong. At a meeting with Air Porce officials, an
argument was made by one officer (who apparently did not know that
GAO had the now two-year-old incriminating memo) that GAO should
be using other (false) data than that used in their report. He was
corrected by another Air Force officer who knew that GAO had both
the accurate data and the 1989 memo. If GAO had not had the memo,
it is evident that the Air Porce would have continued to make false
claims about the RCS of the B-1B using data known to be incor.ect.
As it was, the Air Force's formal final written comments on this
part of the triad study stated in substance that the Air Porce
agreed with GAO's findings on the measurement of the B-lB's RCS and
on the comparison of the B-52 and the B-1B on this issue.
89
The Honorable Lea Aspin
June 28, 1993
Page Two
More recently, incorrect data on the RCS of the B-1B was used
in a letter to me from General John Loh on June 8, 1992. General
Loh stated that "the much older B-52...has a radar signature 100
times larger [than the B-1B]". That, and a questionable
interpretation of the RCS values as giving the B-1B stealth -like
qualities, induced me to sign, with three other Senators, a letter
to our colleagues in support of the B-1B with General Loh's letter
as an attachment. I do not know whether General Loh was in
possession of the correct data when he wrote to me, but the result
in any case was that I was misinformed by the Air Foi_o on the
capabilities of the B-1B. There have been other instances where
I believe the Air Force misled me about B-1B capabilities --
particularly concerning its ability to deliver conventional
ordnance during Desert Storm.
My outrage at being personally misled is dwarfed by my outrage
at the attempt to deceive the Congress as an institution and the
nation at large.
Unfortunately, this is not the only example the GAO provided
at the June 10, 1993 hearing of difficulties obtaining correct data
to evaluate the performance of triad systems. GAO was denied
access to up-to-date data on the reliability of the w-87 warhead
on the Peacekeeper missile on the grounds that it would "pose a
grave threat to national security," despite the fact that earlier
reports containing such data had previously been routinely provided
to GAO. Formal "demand" letters sent by the Comptroller General
in 1990 to then Secretary of Defense Cheney and then-OMB Director
Darman were answered with denials of access to the data, thereby
preventing GAO from corroborating the high estimates of
peacekeeper's warhead reliability.
Mr. Secretary, I hope you are as troubled by this as I am.
For attempts to deceive Congress with false data could also happen
to the highest civilian leadership at the Pentagon. And if you
cannot depend on the accuracy and reliability of the performance
data presented to you, then you will be unable to formulate
reliable plans and policies for the defense of the nation. Deputy
Defense Secretary William Perry, who testified at our hearing, has
assured us that DOD will take steps to prevent such willful
contempt for truth from occurring again. I look for*— rd to any
remedies you may suggest for accomplishing this.
I am enclosing a copy of Ms. Chelimsky's testimony, her post-
hearing letter to me dated June 16, 1993, and General Loh's letter
of June 8, 1992.
90
The Honorable Les Aspin
June 28, 1993
Page Three
I am also sending a copy of this letter to the Acting
Inspector General of DOD, with a request that he undertake an
investigation of the circumstances by which the GAO and the
Congress were supplied false data, and its interpretation, by DOD,
in 1991 and afterward, and provide recommendations regarding
appropriate policy and personnel actions to be taken by DOD.
Sincerely,
(^S John Glenn
Chairman
cc: Deputy Secretary of Defense William Perry
Acting Inspector General Derek vender schaaf
91
'Hnitcd £>t8tcs £>tndtt
I, QMMI IIHQd
mwwim www
wm— mow, oc set lo-ssso
June 28, 1993
Mr. Derek J. Vander Schaaf
Acting Inspector General, DOD
Room 1000
400 Army-Navy Drive
Arlington, VA 22202-2884
Dear Mr. vender Schaaf,
On June 10, I chaired a hearing of the Governmental
Affairs Committee on the General Accounting Office (GAO)
Evaluation of the U.S. Stragetic Modernization Program.
At that hearing, I was shocked to learn, in response to
a question I asked of Assistant Comptroller General Eleanor
Chelimsky, that the Department of Defense (DOD) had provided
falsified data to the GAO concerning the radar cross section
(RCS) of the B-1B bomber.
This deception was discovered by the GAO when the
Strategic Air Command inadvertently delivered a memo
blatantly describing the deception (and showing the correct
data) as part of a package of written DOD comments responding
to a 1989 draft of a GAO report on the B-1B. Subsequent
meetings between the GAO and the Air Force resulted in
changes to the report.
In 1991, GAO was deeply into its study of triad systems
and received draft written comments from the Air Force,
stating in substance that GAO's measurement data and
interpretation thereof on the B-1B was wrong. At a meeting
with Air Force officials, an argument was made by one officer
(who apparently did not know that GAG had the now
two-year-old incriminating memo) that GAO should be using
other (false) data than that used in their report. He was
corrected by another Air Force officer who knew that GAO had
the accurate data and the 1989 memo. If GAO had not had the
memo, it is evident that the Air Force would have continued
to make false claims about the RCS of the B-1B using data
known to be incorrect. As it was, the Air Force's formal
final written comments on this part of the triad study stated
in substance that the Air Force agreed with GAO's findings on
the measurement of the B-lB's RCS and on the comparison of
the B-S2 and the B-1B on this issue.
92
Mr. Derek Vender Schaaf
June 28, 1993
Page Two
More recently, Incorrect data on the RCS of the B-1B was
used In a letter to me from General John Loh on June 8,
1992. General Loh stated that "the much older B- 52... has a
radar signature 100 times larger [than the B-1B]." That, and
a questionable interpretation of the RCS values as giving the
B-1B stealth-like qualities, induced me to sign, with three
other Senators, a letter to our colleagues in support of the
B-1B with General Loh's letter as an attachment. I do not
know whether General Loh was in possession of the correct
data when he wrote to me, but the result in any case was that
I was misinformed by the Air Force on the capabilties of the
B-1B. There have been other instances where I believe the
Air Force misled me about B-1B capabiliities — particularly
concerning its ability to deliver conventional ordnance
during Desert Storm.
My outrage at being personally mislead is dwarfed by my
outrage at the attempt to deceive the Congress as an
institution and the nation at large.
Unfortunately, this is not the only example the GAO
provided at the June 10, 1993 hearing of difficulties
obtaining correct data to evaluate the performance of triad
systems. GAO was denied access to up-to-date data on the
reliability of the W-87 warhead on the Peacekeeper missile on
the grounds that it would "pose a grave threat to national
security," despite the fact that earlier reports containing
such data had previously been routinely provided to GAO.
Formal "demand" letters sent by the Comptroller General in
1990 to then- Secretary of Defense Cheney and then-OMB
Director Darman were answered with denials of access to the
data, thereby preventing GAO from corroborating the high
estimates of Peacekeeper's warhead reliability.
I am troubled that the Congress cannot depend on the
accuracy and reliability of the performance data presented to
it. Deputy Secretary William Perry, who testified at our
hearing, has assured us that DOD will take steps to prevent
such willful contempt for truth from happening again. In
your capacity as the Acting Inspector General of DOD, I am
formally requesting that you undertake an investigation of
the circumstances by which the GAO and the Congress were
supplied false data by DOD, in 1991 and afterward, as well as
its interpretation. In addition, I would like you to examine
why GAO was denied data on the reliability of the W-87. At
the end of your investigation, I further request that you
provide recommendations regarding appropriate policy and
personnel actions to be taken by DOD. Thank you very much
for your cooperation in this matter.
93
Mr. Derek Vander Schaaf
June 28, 1993
Page Three
I am enclosing a copy of Ms. Chelimsky's testimony, her
post-hearing letter to me dated June 16, 1993 and its
attachments, General Loh's letter of June 8, 1992, and my
letter of June 28, 1993 to Secretary Aspin.
Sincerely,
(^ John Glenn
Chairman
CC: Secretary of Defense Les Aspin
Deputy Secretary of Defense William J. Perry
94
GAO
United States
General Accounting Office
Washington, D.C. 20548
Program Evaluation and
Methodology Division
August 6, 1993
The Honorable John Glenn
Chairman, Committee on
Governmental Affairs
United States Senate
Dear Mr. Chairman:
In answer to your letter, I enclose GAO's responses to the
written questions you submitted on behalf of yourself.
Senator Dorgan, and Senator Cochran as follow-up to the
hearing you chaired on June 10 concerning our "Evaluation of
the U.S. Strategic Nuclear Triad."
I want to take this opportunity to express my sincere
appreciation for your work in making the June 10 hearing
possible and for the follow-up work you and your staff have
performed in trying to draw the critical lessons from our
eight reports.
I am also hopeful because of the constructive response the
Department of Defense has adopted with regard to our study.
As you know, Deputy Secretary Perry testified at the hearing
that our findings will be "a very important input" to DOD's
strategic planning. However, I will avoid getting too
enthusiastic until I see some strong studies coming out of
DOD that use valid and reliable data to back up assertions
about weapon system performance and cost.
If you or the other Members of the Governmental Affairs
Committee have any further questions, please call me on
202-512-2900.
Sincerely yours,
Eleanor Chelimsky
Assistant Comptroller General
Enclosure
95
ENCLOSURE ENCLOSURE
Written Questions Submitted bv Senator Glenn
and GAP Responses
Question it Uncertain Costs
The cost of modernizing the triad has uncertainties.
Which systems have the greatest (and the least) cost
uncertainties? How uncertain?
GAP Response
Our assessments of cost uncertainty do not always agree with
those of the Department of Defense (DOD) . A major difference
between DOD's and our cost assessments is that we assess costs
for both the acquisition and deployment phases of the proposed
strategic modernization programs, while DOD typically assesses
only the acquisition phase.
In making the assessments for GAO's evaluation of the
strategic nuclear triad, we analyzed four factors to estimate the
cost uncertainty of a proposed weapon system: (1) the extent to
which the system had already progressed through the DOD
acquisition process and already demonstrated stability in DOD's
own acquisition cost estimates; (2) whether it had completed its
test and evaluation, either with overall success or with some
problems, but with known costs to address them; (3) whether there
was a history of operational deployment to empirically
demonstrate operating and support costs, and (4) whether the
system posed significant technological risk.
With regard to the cost uncertainties of the strategic
programs that are currently deployed or being acquired, we made
the following assessments:
B-lB: The program has encountered numerous performance
problems during both the acquisition and deployment
phases and substantial cost growth has occurred;
nevertheless, the costs to redress the deficiencies,
although high, have been identified based on a
substantial record of testing and deployment.
Accordingly, we assessed cost uncertainty to be low; it
is the cost growth for this system that has been high.
In this latter regard, another GAO report, Strategic
Bombers; Adding Conventional Capabilities Will Be
Complex. Time-Consumina. and Costly (GAO/NSIAD-93-45) ,
finds that the recent assignment of the B-lB to new
conventional roles will add still more costs; but these
also appear to have been reliably identified by DOD.
96
— B-2: During its acquisition phase, the program has
experienced both performance problems and a high level
of instability in DOD's own acquisition cost estimates.
Moreover, there is no deployment history to demonstrate
the accuracy of predicted operating and support costs,
and the program poses a significant level of
technological risk. Accordingly, we assessed cost
uncertainty to be high.
— Ohio-class SSBNs with D-5 SLBMs: Although production of
the last of the 18 SSBNs has not been completed, the
design has been stable, as have DOD cost estimates. In
addition, the system has been deployed for several
years, thus establishing an operations and support track
record. We assessed cost uncertainty to be low.
The analysis of our cost report, U.S. strategic Triad; Costs
and Uncertainties of Proposed Upgrades (GAO/C-PEMD-92-6) , did not
include an assessment of the cost uncertainty of the Minuteman
III (MM III) upgrade and lifeservice extension programs. During
the period in which we were performing our analysis, MM III was
not a proposed modernization. However, based on the assessment
factors outlined above, we would currently assess the cost
uncertainty of the program to be low.
In summary, we found that the B-2 involves the greatest
number of unknowns for any of the weapon systems we examined and
thus presents the highest level of cost uncertainty.
Question 2; Minu»tmflft ttt Modernization. Part I
— Please describe how GAO obtained the figure of $23
billion for Minuteman III modernization, giving a break-
out of the $23 billion budget.
GAO Response
Our report, U.S. Strategic Triad: Final Report and
Recommendations (GAO/C-PEMD-92-8) , relied on data provided by the
Air Force showing that the 30-year lifecycle cost for the MM III
lifecycle extension program was $16.5 billion. We subsequently
found those data to be incorrect: the Air Force's Minuteman III
Life Extension Report, which we received after our report's
publication, made explicit that the $16.5 billion figure was for
a lifecycle cost out to the year 2010, not to the year 2020. To
obtain a 30-year lifecycle cost, we extrapolated the operations
and support costs of the MM III program 10 additional years.
Specifically, we used the Air Force's cost estimate for
various acquisition programs necessary to maintain reliability
and to extend the life of MM III to 2010 ($4.6 billion) and the
Air Force's cost estimate for annual operations and support to
97
the same year ($11.9 billion at $.631 billion per year); to this
we added 10 additional years of operations and support costs at
the same rate of $.631 billion per year ($6.3 billion). A
breakout of the Air Force's cost estimates is provided on page 18
of its Minuteman III Life Extension Report, which we enclose
herewith as appendix I.
Question 2; Minuteman III Modernization. Part II
— Why does GAO recommend examining the decision to
modernize the Minuteman III? What are the advantages
and disadvantages of modernizing the MM III in terms of
capabilities, survivability and other factors?
GAO Response
We stated in our classified report, U.S. Strategic Triad:
Final Report and Recommendations, and in our unclassified
testimony, that we guestion the advisability of funding major
life-service upgrades for the MM III force because the cost-
effectiveness of such an effort is not obvious. There are three
reasons for this: (1) its estimated cost through the year 2020
will be $23 billion, based on Air Force figures; (2) the fact
that a reduced nuclear threat environment exists, both now and in
the foreseeable future; and (3) the likelihood that substantive
modifications would reguire robust flight test programs that
would quickly use up limited test assets.
A re-examination of the MM III life extension program, which
we recommend, might consider the various advantages and
disadvantages that emerged from our analysis. Two major
advantages are that (1) the de-MIRVing of MM III will further
reduce "destabilizing" elements in the triad, and (2) the
lifeservice extension will likely maintain, if not improve, the
relatively high reliability of the MM III force. Four major
disadvantages are that (1) MM III has less military capability
than the D-5 (Mark 5) SLBM in terms of hard target attack
capability; (2) being based in immobile silos, the MM III force
presents an array of easily located targets that an enemy can
choose to attack; (3) confidence in the system's reliability and
accuracy, especially out to the year 2020, will necessarily
deteriorate without new test assets beyond those currently being
contemplated for MM III flight testing; and (4) the MM Ill's $23
billion cost, including significantly greater per warhead costs
than those of other systems (see table 1 below) , may make other
alternatives relatively more attractive.
Question 3; MM III vs. Trident. coat-to-Go. Part I
— What are the costs-to-go in the GAO report on a total
cost basis for the de-MIRVed Minuteman III and the
Trident?
98
GAP Response
Based on the Air Force's Life Extension Report numbers, we
found the total costs-to-go for the de-MIRVed MM III force to be
$23 billion out to the year 2020. We found the total costs-to-go
for the Ohio/D-5 force to be $58 billion, assuming a 30-year
lifecycle for each SSBN. However, because some Ohio class SSBNs
were deployed in the 1980s and will be retired before 2020, we
calculated the average lifecycle cost-to-go for the entire force
of 18 Ohio SSBNs to be 25.6 years from the year 1992 forward. To
take the difference in lifecycle assumptions for the two systems
more fully into account, we calculated cost per warhead per year
for the MM III and Ohio/D-5 systems. These calculations are
shown in table 1.
Table 1: Costs-to-Go per Warhead per Year for Minuteman III and
Ohio/D-5 (with 1 warhead per Minuteman III and 4 per
D-51
System
Assumed
Life-
cycle
From
1992
Life-
cycle
Cost-
to-Go
Number
of
Warheads
Dollars
per
Warhead
Dollars
per Year
per
Warhead
MM III
Compared
to
Ohio/D-5
MM III
18 years
to the
year
2010
$16.5
billion
500
$33
million
$1.83
million
40%
higher
than
Ohio/D-5
MM III
28 years
to the
year
2020
$23
billion
500
$46
million
$1.64
million
25%
higher
than
Ohio/D-5
Ohio/
D-5
25.6
years on
average
for each
SSBN
$58
billion
1728
$33.6
million
$1.31
million
Note that these estimates do not take into account various
differences between the MM III and Ohio/D-5 systems, such as the
essential invulnerability of the SSBNs at sea compared to the
known (attackable) locations of ICBM silos, the greater accuracy
of the D-5 SLBM, or different likely alert rates for ICBMs and
SLBMs under a variety of different scenarios. The analyses that
incorporate these factors are classified.
99
Question 3: MM III va. Trident. Cost-to-Qo, Part II
— Do these costs take into account (1) the retention of
C-4s in the Pacific fleet, (2) the increase of MM III
costs from $16 billion to $23 billion, (3) the lifetimes
of the Trident SSBNs, and (4) the decommissioning costs
of nuclear weapons and nuclear reactors?
GAP Response
The calculations given in table 1 above make the following
assumptions:
(1) The lifecvcle costs estimated for all 18 Ohio SSBNs are
based on an assumption that each would carry 24 D-5
missiles with 8 warheads each. (Retention of C-4
missiles, even including lifeservice extension for
them, if necessary, would likely be less costly.)
Estimates in table 1 for dollars per warhead, however,
are based on 4 warheads per missile, to comply with the
1750 SLBM warhead ceiling of the START II treaty.
Overall costs (or savings) to download the SLBM force
from 8 to 4 warheads per missile are not included.
(2) As shown, MM III lifeservice extension costs to the
year 2010 ($16.5 billion) and to the year 2020 ($23
billion) are taken into account.
(3) We used an average remaining lifeservice of 25.6 years
from the year 1992 for the entire Ohio SSBN force.
This calculation assumes a lifeservice of 30 years from
deployment for each Ohio SSBN.
(4) As we point out in our report, U.S. Strategic Triad;
Costs and Uncertainties of Proposed Upgrades, nuclear
warhead and/or power plant disposal costs for either MM
III or the Ohio/D-5 force and the scrap costs (and
value) of non-nuclear components of systems were not
available from the Department of Defense and are not
included.
Question 3: MM III v. Trident. Coat-to-Go. Part III
— Which system is more capable, the MM III or the Trident
II D-5 with W-88 warheads? Which one is capable of
destroying "hard-targets?"
GAP Response
At the time we performed our analysis, the MM III was not a
proposed upgrade for the land leg of the triad. Thus our
comparison for the D-5 SLBM was not to Minuteman but to the then
100
proposed upgrade Peacekeeper ICBM. We performed this comparison
using the following measures of effectiveness: (1) speed and
reliability of communications to command authority; (2) time to
target, especially for time urgent targets; (3) pre-launch
survivability against pre-emptive attack; and (4) lethality to
enemy targets, including accuracy, warhead yield, and
reliability.
On the measures of speed and reliability of communications
and on time to target, we found Peacekeeper and the Ohio/D-5
system with W-88 (Mark 5) warheads to be essentially equal in
terms of actual operational effectiveness. For pre-launch
survivability, we found that SLBMs on submerged SSBNs possess a
clear advantage over ICBMs in easily locatable silos. And we
found the D-5 Mark 5 SLBM to be equivalent to the Peacekeeper
ICBM on the measure of capability to destroy hard targets.
In contrast, MM III is considerably less accurate than
either the Peacekeeper or D-5/Mark 5, and hence would be much
less lethal against hard targets. On other measures, such as
communications speed and system reliability, MM III is
essentially the equal of the two other systems, while sharing the
same deficiency as Peacekeeper with regard to easy locatability,
given its basing mode in fixed silos.
In sum, the D-5 with Mark 5 (W-88) warheads would be
considerably more capable with regard to hard targets than the
MM III.
Question 4t ICBM-silo vulnerability
— Has your analysis on silo vulnerability taken into
account the effects of (1) fratricide on second and
subsequent warheads and (2) the lack of an ability to
practice such an attack? Generally, how would these two
effects change your estimates of silo kill probability?
GAP Response
Our report, V.S, Strategic Triad; ICBM Vulnerability
(GAO/C-PEMD-92-2) , found that the claimed increase in
vulnerability of U.S. ICBMs in their silos due to improved Soviet
missile accuracy had been greatly overstated. The specific
calculations we employed in this study followed a methodology
provided to us by the Department of Defense, using intelligence
community data to incorporate the performance of the most lethal
Soviet ICBMs.
As our study noted, the former Soviet Union would have faced
a number of major uncertainties about their missiles' performance
in planning any attack on the U.S. Many of these uncertainties
could not have been confidently resolved short of actual nuclear
101
conflict. Since no computer model can adequately simulate all
actual attack conditions — and since we were in part addressing
the claim that U.S. ICBMs were highly vulnerable to massive
Soviet attack — we chose to examine U.S. ICBM losses assuming
Soviet systems of maximum effectiveness. This meant no. Soviet
losses of incoming warheads to fratricide. The number of
incoming Soviet warheads in our analysis was reduced only by the
system reliability factor estimated by the U.S. intelligence
community for particular Soviet missiles. Soviet warheads were
further assumed to detonate at the altitude that would cause the
greatest damage to U.S. silos. And we assumed that no U.S. ICBMs
were launched on warning or even under attack — both of which
would reduce U.S. losses.
In effect, this analysis embodied the so-called "worst-case"
scenario from the U.S. point of view. But even in this case, we
calculated that more U.S. ICBMs would survive than had been
claimed. Further, it is readily apparent that if fratricide and
test unrealism were quantifiable in some reliable manner, the
impact on our calculations would have been to increase the number
of surviving U.S. silos.
The point here is that any model that could reliably predict
U.S. ICBM survivability — by taking into account uncertainties for
the attacker that are not currently measurable — would demonstrate
less vulnerability than the "worst case" scenario. Further,
accounting for fratricide and test unrealism, if that could have
been done in some reasonably rigorous way, would have had the
effect of augmenting the number of U.S. ICBM silos that could
ride out a former Soviet ICBM attack, survive, and retaliate.
Question 5; Air Defense Estimates
— Are the 1976 CIA Team B projections for Soviet air
defenses consistent with those used by the DOD in the
late 1970s and early 1980s?
GAP Response
We did not go back to any intelligence sources, such as the
1976 Team B projections, that pre-dated the Defense Department's
justifications to Congress for acquiring both the B-1B and the
B-2. One of the major baseline sources of our analysis of the
size of Soviet strategic air defenses and of the accuracy of
intelligence community projections that were made for them was
the October 1981 Joint OSD/Air Force Bomber Alternatives Study.
It was this study that presented the DoD justifications to
Congress for the B-1B and the B-2, in very large part on the
basis of claimed increases in the number and effectiveness of
Soviet strategic air defenses. The study's estimates were based
on the latest threat analysis from the U.S. intelligence
community available at the time. The specific sources were
official intelligence community assessments, such as the Defense
Intelligence Projections for Planning ("DIPPs") which are
coordinated throughout the various agencies of the intelligence
community.
102
Written Questions Submitted bv Senator Doraan
and GAP Responses .
Question 1; The Air Force's proposed Minuteman Life Extension
Program involves an acquisition cost of $4.6B for modifications.
Air Force cites a total cost-to-go of $16B based on the $4.6B of
acquisition programs and $11. 9B of operations and support for the
1992-2010 timeframe. You have chosen the timeframe to the year
2020 as a basis for comparison. In your cost conversion, you
appear to have extrapolated without first removing the
acquisition cost. Hill you please make the necessary correction?
GAP Response
No correction is required because we did not include
acquisition costs in the 2010-2020 estimate, having already
included them in the 1992-2010 estimate. The calculation I
presented in my testimony — $23 billion for the 30-year lifecycle
cost for MM III — is not the same as the number ($16.5 billion)
given in our report, U.S. Strategic Triad; Final Report and
Rec9mm?ndat ions (GAP/C-PEMD-92-8) . This report relied on data
provided by the Air Force showing that the 30-year lifecycle cost
for the MM III program was $16.5 billion. We subsequently found
those data to be incorrect: the Air Force's Minuteman III Life
Extension Report, which we received subsequent to the publication
of our series of reports, made clear that the $16.5 billion
figure was only for a lifecycle out to the year 2010, not to the
year 2020. To obtain a 30-year lifecycle cost, we extrapolated
the operations and support costs of the MM III program 10
additional years.
Specifically, we used the Air Force's cost estimate for
various acquisition programs necessary to maintain reliability
and to help extend the life of MM III to 2010 ($4.6 billion). We
also adopted the Air Force's cost estimate for annual operations
and support to the same year ($11.9 billion at $.631 billion per
year) ; these costs are needed in addition to the acquisition
costs cited above for normal operations and maintenance and
general facility support. To this $11.9 billion for operations
and support, we added 10 additional years of operations and
support costs at the same rate of $.631 billion per year ($6.3
billion). ($4.6 billion + $11.9 billion + $6.3 billion - $22.8
billion; statistical rounding gives the figure of $23 billion.)
Question 2; Your estimate of the life cycle cost-to-go for the
Trident system is $58B based on a timeframe out to the year 2020.
How did you compute the annual operational and support (O&S) cost
of a Trident boat? Did you include cost of operating shore bases
such as Bangor and Kings Point? To be more specific, did you
include, for each base, costs of maintaining roads, grounds and
buildings; medical facilities; dining facilities; dormitories;
103
salaries of indirect support people; recreational activities?
(In calculating ICBM costs, all of these elements are included.)
Including all these elements, what is the annual O&S cost of a
Trident submarine and its missiles?
GAP Response
As specified in our report, U.S. Strategic Triad; Costs and
Uncertainties of Proposed Upgrades (GAO/C-PEMD-92-6, see pages
14-15, and 32-33), our Ohio/D-5 lifecycle cost analysis does
include both the direct and indirect costs of operating the
Trident system-related facilities at Bangor and Kings Point.
Specifically, our analysis of the costs to operate and support
the Ohio/D-5 system includes the following elements:
(1) Military construction costs for the Trident missiles
and submarines ($1.2 billion);
(2) General SSBN operating costs, including the costs to
convert C-4 capable SSBNs to the D-5 missile, missile
industrial facility maintenance, general support costs
(such as base infrastructure costs like housing,
chapels, and theaters) , and ongoing investment costs
(totaling $40.7 billion);
(3) Fleet ballistic missile support ($.7 billion); and
(4) Headquarters support ($3.7 billion).
The total of these operations and support (O&S) costs is
thus $46.3 billion.
As our report states on page 14, our cost estimates do not
include the disposal or scrap costs, and value, of materials at
the end of the system lifecycle. DOD officials informed us that
such cost estimates were not available, and for that reason we
could not include them.
Based on the acquisition and operations and support costs
yet to be expended as of the end of fiscal year 1990, we
calculated the total costs-to-go for the Ohio SSBN/D-5 SLBM
system to be $58 billion. Of course, to calculate the system's
cost-to-go on a more current basis — that is, as of the end of
fiscal year 1992, which is the date of the Air Force's Life
Extension study estimating costs to the year 2010 — would require
a downward adjustment of the $58 billion figure.
Question 3; A certain number of Trident boats will age out
before the year 2020. In your cost-to-go, did you include boat
replacement costs? In cases where submarine nuclear reactors
will require re-coring, did you include the relevant costs?
10
104
GAP Response
We did not calculate costs for any follow-on system to
replace either Ohio class SSBNs or D-S missiles. Our lifecycle
cost analysis assumed a 30-year lifecycle for each Ohio class
SSBN. Because a portion of the fleet vas deployed in the 1980s
and is scheduled for retirement before the year 2020, we have
calculated the average lifecycle to go for the entire force of 18
Ohio SSBNs to be 25.6 years from the year 1992 forward, and we
have calculated a cost per warhead per year for the Ohio/D-5
system. He have performed a similar calculation for the MM III
lifeservice extension program, using both the Air Force's $16
billion estimate for service to the year 2010 and our
extrapolation of operation and support costs to the year 2020.
The results of these calculations are shown in table 1. (This
table, already presented earlier, is repeated here for the
reader's convenience.)
Table 1: Costs-to-Go per Warhead per Year for Mjjmfcemaj] III and
Ohio/D-5 fwith 1 warhead per Mlnuteman III and 4 per
D-5)
System
Assumed
Life-
cycle
From
1992
Life-
cycle
Cost-
tO-GO
Number
of
Warheads
Dollars
per
Warhead
Dollars
per Year
per
Warhead
MM HI
Compared
to
Ohio/D-5
MM III
18 years
to the
year
2010
$16.5
billion
500
$33
million
$1.83
million
40%
higher
than
Ohio/D-5
MM III
28 years
to the
year
2020
$23
billion
500
$46
million
$1.64
million
25%
higher
than
Ohio/D-5
Ohio/
D-5
25.6
years on
average
for each
SSBN
$58
billion
1728
$33.6
million
$1.31
million
Note that these estimates do not take into account various
differences between the MM III and Ohio/D-5 systems, such as the
invulnerability of the SSBNs at sea compared to the known
(attackable) locations of ICBM silos, the greater accuracy of the
11
105
D-5 SLBM, or different likely alert rates for ICBMs and SLBMs
under a variety of different scenarios.
The calculations above make the following assumptions:
(1) The lifecvcle costs estimated for all 18 Ohio SSBNs are
based on an assumption that each would carry 24 D-5
missiles with 8 warheads each. (Retention of C-4
missiles, even including lifeservice extension for
them, if necessary, would likely be less costly.)
Estimates for dollars per warhead in table 1, however,
are based on 4 warheads per missile, to comply with the
1750 SLBM warhead ceiling of the START II treaty.
Overall costs (or savings) to download the SLBM force
from 8 to 4 warheads per missile are not included.
(2) As we point out in our report, U.S. Strategic Triad:
Costs and Uncertainties of Proposed Upgrades, nuclear
warhead and/or power plant disposal costs for either MM
III or the Ohio/D-5 force and the scrap costs (and
value) of non-nuclear components of systems were not
available from the Department of Defense and are not
included.
(3) These data do not include a potential additional $1.9
billion being considered by the Air Force to improve
the inertial measurement unit of the MM Ill's guidance
system for the purpose of improving accuracy.
(4) Costs for the Ohio class SSBNs do include all of the
Navy's estimates for operating and maintaining the
system, such as any nuclear power plant re-corings
scheduled before the year 2020. Only the costs of
disposing of the used core, once replaced, were not
available from the Navy and are not included.
Question 4: The C-4 missiles in the Pacific Ocean submarines
will begin to age out before the year 2020. We understand that
the Navy wants to backfit with D-5 missiles. Have you included
the cost of backfitting, including the costs of the boat
modifications needed to accommodate the backfit?
GAP Response
Our lifecycle cost estimates do include the costs of
backfitting the C-4 Ohio class SSBNs with D-5 missiles, involving
the costs of both the new missiles and the necessary
modifications to the submarines themselves.
12
106
Written Questions Submitted by Senator Cochran
and GAP Responses
Panel 1; Eleanor Chelimsky. Assistant Comptroller General for
Program Evaluation and Methodology GAP
Ms. Chelimsky, I'd like to thank you for taking the time to
appear before this committee to discuss a serious topic which so
directly affects our nation's security. I'd also like to commend
the General Accounting Qffice for studying the cost effectiveness
of the various legs of the triad. Certainly, any serious
discussion of how to structure and arm our nation's military must
examine cost effectiveness — GAQ's yardstick in this study — as one
of the many components considered.
Question 1: I am, however, concerned that this study focuses so
heavily on the question of which upgrades to the Triad are most
cost effective. The cost of a program is something my colleagues
and I take very seriously, but does this approach not limit the
flexibility of a U.S. response?
GAG Response
Two points are relevant here: first, we did not perform one
analysis of cost-effectiveness, we performed several; and second,
we did not perform only cost-effectiveness analysis.
Concerning the first point, we measured cost in a variety of
ways because we believe that no single measure provides a
definitive picture even of the cost issue taken alone.
Accordingly, we assessed: acquisition costs (the measure most
frequently cited by DCD) ; total lifecycle costs (combining
acquisition costs [RDT&E and procurement] with all available
forms of operating and support costs) ; sunk lifecycle costs, and
lifecycle costs yet to be expended ("costs-to-go") . Using these
different cost measures, we then performed various cost
effectiveness calculations on a cost per warhead basis. In this
regard, we calculated costs per deployed warhead for each
proposed modernization system, and we performed cost per arriving
warhead calculations employing classified DOD assumptions
concerning alert rates, reliability, and survivability. We
performed these latter cost per arriving warhead calculations for
two different scenarios: surprise Soviet attack, and attack
under generated alert, that is, with strategic warning.
In addition to performing these cost analyses, we assessed
the effectiveness of each proposed modernization plan and of
several previously deployed systems using seven different
13
107
measures of effectiveness.1 Thus, from a methodological
viewpoint, our analysis of cost effectiveness issues was quite
comprehensive .
Concerning my second point above, GAO never imagined that
cost effectiveness could be the sole basis for comprehensively
evaluating the strategic triad — although it is an important
basis — because it fails to capture several critical concepts and
elements. For precisely that reason, we did not limit either our
conceptualization of the issues or our specific analyses to cost
effectiveness. We examined many other aspects of the triad and
specifically performed the following additional analyses, all of
which can be found in the evaluation's final report (GAO/C-PEMD-
92-8) :
The impact of both existing strategic systems and
proposed modernizations on arms control considerations,
and the impact of specific provisions of the START I and
II treaties on the systems and their upgrades (pp. 200-
201) .
The interactions of various systems within legs of the
triad, such as whether the air-leg's combination of
penetrating bombers and cruise missile carrying aircraft
enhanced the ability of either system to achieve mission
objectives (pp. 187 and 194) .
Interactions among the three legs of the triad,
including whether and how dissimilar performance
characteristics — both strengths and weaknesses — of the
systems in the different legs were complementary or
merely redundant. For example, we assessed the
feasibility and likely effectiveness of attacks on each
of the very different basing modes of the three legs of
the triad (bomber bases, ICBH silos, and SSBNs at sea)
employing different attack scenarios — either
1The seven measures of effectiveness we employed were the
following: (1) survivability against both offensive and
defensive threat systems, for both platforms and weapons (for
example, submarines and their ballistic missiles; bombers and
their cruise missiles) ; (2) delivery system performance (that is,
accuracy, range, and payload) ; (3) warhead yield and reliability
(that is, the probability that the warhead will detonate as
intended); (4) weapon system reliability (that is, the combined
reliability of all the component processes from platform launch
to warhead detonation) ; (5) flexibility across a number of
dimensions, including retargeting and recall; (6) communications
(for example, connectivity between command authority and
platforms) ; and (7) responsiveness (that is, alert rate and time-
to-target) .
14
108
simultaneous launch of attack systems or simultaneous
detonations of the warheads employed (pp. 190-191) .
— The likely cumulative or synergistic impact of the three
legs of the triad on Soviet nuclear force planning and
strategy. Specifically, we first assessed whether, if
deployed, each proposed U.S. strategic system would
have: (1) provided the U.S. with a technological hedge
against the capabilities of defensive or offensive
Soviet systems; (2) complicated Soviet planning, (3)
helped dilute Soviet material resources, and/ or (4) had
a positive or negative effect on the U.S. ability to
respond flexibly to a range of Soviet attack scenarios
(p. 199). Second, we examined whether existing U.S.
systems and proposed upgrades, taken together, enhance
or detract from U.S. deterrence of Soviet attack and
whether the performance of new and currently deployed
U.S. systems can be considered adequate to support an
effective deterrence strategy. For example, while we
found that adding mobility to ICBM systems would,
indeed, increase their survivability, we found that a
U.S. silo hardening program would have similarly
increased silo-based ICBM survivability. However, based
on other analyses — in this case, threat analysis — we
found that neither program was necessary to preserve
either the survivability of U.S. retaliatory
capability — including that of ICBMs — or deterrence (pp.
195-196) .
It is important to recall that my statement at the
Committee's June 10th hearing presented only highlights of our
evaluation and that our nearly thousand-page, 8-volume study
considered many other aspects of the triad (as discussed above)
that could not be covered in the time available.
Question 2 : Oo you believe cost effectiveness to be the
paramount consideration in upgrading the legs of the Triad?
GAP Response
As already noted in my answer to your first question, we
believe that cost-effectiveness analysis (such as cost-to-go per
arriving warhead under strategic warning) , based on valid and
reliable data, is one of several forms of analysis that is needed
to perform a comprehensive evaluation of a complex matrix of
systems such as the strategic triad. In the particular case of
the triad, however, such analysis takes on exceptional importance
because it is more than 30 years since this kind of work has been
performed. On balance, I would suggest that a thorough
assessment — such as the one already conducted by GAO, or the
Defense Department's current "bottom up" review — would be
incomplete if it consisted only of a cost-effectiveness analysis,
15
109
but that it would also be Incomplete — like so very many other
assessments of the past — if it did not include analysis.
Question 3: In the past, we were dealing with the Soviet Union
within the framework of deterrence. This framework posited that
deterrence would "work" as long as both sides were risk averse,
rational actors. DCI Jim Woolsey, in his confirmation hearing a
few months back, stated "He have slain a large dragon, but we now
live in a jungle filled with a bewildering variety of poisonous
snakes, and in many ways the dragon was easier to keep track of."
Is the GAO so sure of the future that you are willing to state
that we need cost effective forces, which is where your report
leads us, instead of flexible forces with a spectrum of
capabilities?
GAO Response
As requested by the Congress, our review was of U.S.
strategic systems in the context of the U.S. relationship with
the former Soviet Union. During the course of our work, we found
that the U.S. possession of flexible strategic nuclear forces,
able to respond to a range of scenarios with a spectrum of
capabilities, was an essential element of an ability to deter and
if necessary respond to attack, even from the unitary "dragon"
that Director Woolsey referred to in his confirmation hearing.
Specifically, we evaluated flexibility in two different
ways. First, we analyzed the ability of each individual
strategic system to operate in a flexible manner: that is, we
assessed the extent to which, if at all, systems could be
recalled, retargeted, or could attack a variety of target types
under differing circumstances. Second, we assessed the extent to
which the overall triad gave the President a variety of different
retaliatory options: that is, whether the capabilities in the
various legs of the triad were in fact complementary.
In effect, our reports analyze the triad systems across a
number of dimensions that are directly relevant, not just to the
Cold War, but also to evaluating the usefulness of these systems
in the present and future. It is readily apparent that the
recallability of bombers gives them a performance flexibility
that is considerably greater than ballistic missiles, and that
this could be useful in conventional post-Cold War crises.
At the same time, it is clear that the focus of our work — as
requested — was the capability of U.S. strategic systems vis-a-vis
the then-existing principal threat, the Soviet Union. Analysis
of the applicability of U.S. strategic forces to dealing with,
say, third-world states possessing nuclear stockpiles of a dozen
weapons or less, and exclusively medium-range missiles, was
outside the scope of our study.
16
110
GAO's evaluation does not state that cost-effectiveness
should take precedence over other dimensions or measures in
deciding what strategic systems offer the most advantages.
Indeed, our reports make clear that there are multiple important
non-cost dimensions to these systems, such as flexibility or arms
control. The analysis presented in the eight reports show the
desirability for the United States to retain strategic forces
that are flexible and that possess a spectrum of capabilities to
cope with a variety of potentially unforeseen threat scenarios.
However, it is also important that these systems be cost-
effective, given that it makes little sense to spend money on new
systems to obtain capabilities that already exist in our military
inventory, or that are designed to meet non-existent or vastly
overestimated threats.
The short answer to your question, then, is that we believe
our report leads not to the conclusion that we need cost-
effective forces, but that whatever forces we need should be
subjected to cost-effectiveness (and other) analysis on a fairly
regular basis.
Question 4; When the GAO began this study in 1990, Peacekeeper
Rail Garrison and the Small ICBM were ongoing programs, and we
were projecting a much larger B-2 buy than we are today. Given
the demise of the first two programs, and the evisceration of the
B-2 program, how valid does your study remain?
GAO Response
It is correct that we were asked to address certain systems
that are no longer being developed or procured. However, our
reports also address eight systems that are currently under
development, being flight tested, or currently deployed — and that
will remain viable for many years to come.2
While our studies produced numerous findings and
recommendations, many of them classified, we believe three of
these merit serious attention in the post-Soviet world. These
are:
1. On balance, the evidence shows the sea- leg to be the
strongest, most cost-effective component of the triad
under a range of scenarios. We believe this conclusion
has special importance as we enter a post-Soviet period
of uncertainty and major domestic budgetary pressures.
I would remind you that Deputy Secretary Perry agrees
with us on this point: he commented at the Committee's
June 10th hearing that in his view also, the Trident
ZThese systems include the B-S2H, B-1B, B-2, ALCM, ACM, MM III,
Ohio SSBN, and D-S SLBM.
17
Ill
system is the "dominant leg of the triad" and that "we
could get by very nicely with much smaller forces and
[even] a single leg."
Because strategic bombers are flexible and stabilizing,
they add a critically important character to the overall
nuclear force. Again, we believe this finding has
significant implications for our post-Soviet force
structure .
Finally, the multiple flaws and failures we found in
areas like threat forecasting, testing, and reporting
are, if anything, more important for the future than the
past. As you know, our data showed that Soviet
threats — whether to our land, sea, or air leg weapon
systems — were consistently inflated beyond what was
actually the case; testing of the U.S. systems was
frequently truncated or unrealistic or both; and
reporting on a variety of subjects was frequently
incomplete, pessimistic with regard to mature systems,
and optimistic about the likely cost and performance of
new ones. We also found a dearth of the comparative
evaluations needed to show whether a proposed system is
justified in terms of three things: the threat it
faces, its performance capabilities vis-a-vis other
systems, and its relative costs.
These flaw--, and omissions are not, as some have claimed,
the characteristics of an acquisition strategy that is
exercising prudence by preparing for a "worst case
scenario." Indeed, it can never be prudent to have
imprecise or distorted information, either about U.S. or
enemy capabilities. On the contrary, we would argue
that such a situation is inherently imprudent . because
it allows unhappy surprises in time of war and costly
consequences for the U.S. budget. Overall, the failures
of information we found were serious, pervasive, and
persistent over the past 10-15 years; one measure of the
validity of our study will certainly be the degree to
which it helps us avoid these failures in the future.
18
J
112
APPENDIX I
APPENDIX I
Totals Through 2010
The total cost of sustaining the Minuteman III ICBM force through
FY 2010 is estimated to be $16.5 billion. An additional $1.9
billion is being considered to improve the inertial measurement
unit of the Guidance and Control Unit.
These costs are summarized in Table 2.
Minuteman III Life Extension Costs Through FY 2010
(FY 1992 Constant Dollars in Millions)
Stage 1
Stage 2
Stage 3
PSRE
Guidance
Rivet MILE
Non-Missile Equipment
REACT
— -c-
MEECN
Others
Operations & Support ($631M per year)
Military Personnel
Depot Maintenance
Sustaining Investment
Sustaining Engineering
other Direct Costs (SAC)
Indirect Costs
Total FY92-2010
FY 1992-2010
$657
$639
$432
$250
$1,406
$265
$310
$346
$113
$72
($11,989)
$3,819
$931
$380
$912
$3,781
$2,166
$16,479
113
INSERT FOR THE RECORD
SENATE
APPROPRIATIONS COMMITTEE
HEARING DATE
Jun 101993
TRANSCRIPT PAGE NO.
HOUSE
SENATE
ARMED SERVICES COMMITEE
LINE NO.
HOUSE
OTHER
Government Affairs
Question 1
DOD ANALYSIS ON THE TRIAD
Senator Glenn: GAO states that they found that no DoD
evaluation had examined U.S. strategic forces in this
comprehensive way for at least three decades, and that GAO was
concerned to find little or no prior recent effort by DoD to
evaluate comprehensively the relative effectiveness of similar
weapons systems. — Do you agree with the GAO's assessment? What
kinds of inter-leg comparisons did DoD carry out on the Triad?
Dr. Perry: The Department strongly disagrees with the GAO
assessment, which is based on their study that focuses primarily
on cost and fails to properly recognize other critical measures
of effectiveness appropriate for decisions regarding the
acquisition and deployment of nuclear forces.
The
part, has
mutually
leg hedge
beginning
or did ma
quality,
acquisiti
Triad, and the national strategy of which it has been a
been successful largely because the different legs are
reinforcing. The capabilities and attributes of each
against perceived weaknesses of the others. From the
, our analysis of the weapon systems options, that could
ke up the Triad has been comprehensive and of high
and was utilized for important decisions regarding the
on and operations planning.
Analysis has been conducted throughout the Department (plus
elsewhere in the Executive Branch) at various levies, from
different perspectives (e.g., concept designs, acquisition
alternatives, arms control), and on the basis of up-to-date test
and design data. The Strategic Command (previously the JSTPS and
Strategic Air Command), the Services (including their
headquarters and system program offices), the Joint Staff, and
OSD all have had dedicated offices which conduct evaluations on
the cost and combat and deterrent effectiveness of our strategic
forces. The records provide for countless presentations to the
Congress over the years detailing costs, effectiveness, and
performance requirements during the Department's budget requests
regarding such undertakings as the Strategic Modernization
Program, Peackeeper basing modes, cruise missile programs,
submarine-launched ballistic missile programs, Small ICBM, the
1986 Bomber Study, and many more.
Today, on the basis of updated intelligence estimates and
system performance assessments, the U.S. Strategic Command
continually analyzes the effectiveness of our nuclear forces for
changing conditions and the full range of plausible scenarios.
Under the Unified Command Plan, USSTRATCOM is tasked with
formulating war plans, future strategic nuclear force structures,
and systems analyses of the capabilities of current and planned
strategic nuclear forces. Since assuming the role as the single
voice for strategic systems on June 1, 1992, USSTRATCOM has
conducted studies about our current capabilities.
114
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
V
SENATE
HEARING DATE
Jun 1C1993
TRANSCRIPT PAGE NO.
LINE NO.
INSERT NO.
Question 1
DoD ANALYSIS ON THE TRIAD (CONTINUED)
The Joint Staff has a directorate dedicated to conducting
analysis, net assessments, and evaluations of military forces,
plans, programs, and strategies for the Chairman of the Joint
Chiefs, and support of the Secretary and the President. The
Force Design Division and the Nuclear Forces Analysis Division
within this directorate are dedicated to analyzing U.S. strategic
forces. Their efforts include developing force structure
alternatives, conducting tradeoff analyses, cost and budget
analysis independent of the Services, and preparing comparative
analyses and net assessments of weapon systems.
Within the staff to the Secretary of Defense (Policy,
Acquisition, and Program Analysis and Evaluation) ongoing
oversight and analysis of various aspects of our strategic forces
evaluate alternative force structures as well as individual
system performance. These offices each work with their
counterparts within the Joint Staff, the Services, and the
Unified and Specified Commands to investigate issues of interest
based on changing constraints.
I hope this discussion gives you a clearer picture of the
type of analysis and oversight the Department conducts. It is my
judgment that the Department has conducted appropriate analysis
over the years with respect to our strategic forces.
Senator Glenn: If DoD failed to carry out the studies which
compared land-, sea-, and air-based weapons, to what do you
attribute this lack of analysis?
Dr. Perry: The United States Government in general, and the
Department of Defense in particular, has not failed to properly
analyze our past, current, and potential forces. Decisions
regarding our strategic forces have responded to, and in many
cases have exploited, changing threats, technology, and the
results of arms control negotiations. In each case, extensive
and independent analyses and assessments were conducted by the
Department and Services.
Senator Glenn: Which part of the Pentagon should do such
analysis? The Defense Acquisition Board? The Defense Science
Board? The Joint Chiefs?
Dr. Perry: USSTRATCOM now has primary responsibility for
strategic nuclear forces to support the national objective of
strategic deterrence, is the primary military voice for strategic
forces issues, and shares that responsibility with the
appropriate components within the Office of the Secretary of
Defense, the Joint Staff, and the Services.
115
INSERT FOR THE RECORD
HOUSE
SENATE
APPROPRIATIONS COMMITTEE
HEARING DATE
Jun 10 1993
TRANSCRIPT PACE NO.
SENATE
ARMED SERVICES COMMITEE
INSERT NO.
HOUSE
SENATE
OTHER
Government Affairs
Question 2
U.S. Strategic TRIAD:
U.S. Strategic TRIAD:
Senator Glenn: Areas of Disagreement: GAO stated that
«»*?£ i s^emati-c disparities between estimates or claims
";f°"the likely cost and performance of upgrades, the actual
performance of current systems, or the likely offensive or
defensive threats to these systems from the former Soviet Union.
be^nn^rk^fnd^oD^0^ ' —«***- °f th* disagreements
n'SlJV*'' V't. f ollowin9 li3t Provides references to the
extensive documentation on GAO positions and the DoD responses:
Final Report and Recommendations,
Dated September 25, 1992, GAO/C-
PMED-92-8 (OSD Case 8936-X)
DoD Response to Draft:
Appendix V, pp. 235-274
Current Status, Modernization Plans,
and Doctrine of British and French
Nuclear Forces, Dated September 25,
1992, GAO/C-PEMD-92-7 (OSD Case
8672X)
DoD Response to Draft:
Appendix II, p. 42
Costs and Uncertainties of Proposed
Upgrades, Dated September 25, 1992,
GAO/C-PEMD-92-6 (OSD Case 8693-X)
DoD Response to Draft:
Appendix II, pp. 67-78
Strategic Relocatable Targets,
Dated September 25, 1992, GAO/C-
PEMD-92-5 (OSD Case 8637-X)
DoD Response to Draft: Appendix II,
pp. 48-58
U.S. Strategic TRIAD:
U.S. Strategic TRIAD:
U.S. Strategic TRIAD:
Modernizing Strategic Bombers and
Their Missiles, Dated September 25,
1992, GAO/C-PEMD-92-4 OSD Case
8727-X)
DoD Response to Draft: Appendix V.
pp. 142-164
116
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
J
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PACE NO.
LINE NO.
INSERT NO.
Question 2 (Com)
(Cont)
U.S. Strategic TRIAD:
U. S. Strategic TRIAD:
A Comparison of ICBMs and SLBMs,
Dated September 25, 1992,
GAO/C-FEMD-92-3 (OSD Case 8802-X)
DoD Response To Draft: Appendix IV,
pp. 156-186
ICBM Vulnerability, Dated September
25, 1992, GAO/C-PEMD-92-2 (OSD Case
8801-X)
DoD Response to Draft:
II, pp. 54-64
Appendix
U.S. Strategic TRIAD:
ICBM Vulnerability of Strategic
Ballistic Missile Nuclear
Submarines, Dated September 25,
1992, GAO/C-PEMD-92-1 (OSD Case
8446-X)
DoD Response to Draft:
III, pp. 67-82
Appendix
117
INSERT FOR THE RECORD
HOUSE
SENATE
APPROPRIATIONS COMMITTEE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
HOUSE
SENATE
ARMED SERVICES COMMITEE
INSERT NO.
OTHER
Government Affairs
Question 3
THE CLINTON-ASPIN NUCLEAR FORCE
Senator Glenn: In what ways will the nuclear force
structure of the Clinton Administration differ from that of the
Bush Administration?
Dr. Perry: The Depar
structure review, known as
conducting further review
strategy, doctrine, force
supporting infrastructure)
Defense Planning Guidance,
comment on the results of
review is complete we will
tment has conducted an extensive force
the Bottom Up Review, and is
of our nuclear posture (policy,
structure, command and control, and
consistent with the recently issued
It would be premature for me to
that study now; however, when the
provide the results to the Congress.
118
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
J
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
UNE NO.
INSERT NO.
Question 4
UNCERTAIN COSTS
Senator Glenn: The cost of modernizing the Triad has
uncertainties. Which systems have the greatest (and the least)
cost uncertainties? How uncertain?
Dr. Perry: It is not possible to quantify the precise degree
of cost uncertainty with any acquisition program. However, a
program will have more uncertainty when there is more engineering
design and development to accomplish.
On this basis, the Trident D-5 program could be judged to
have the least cost uncertainty, because it has completed
engineering design and development, and is in a mature production
phase of procurement.
The B-2 program is now completing engineering design and
development. It is approximately thirty-five percent through
flight test, and is completing its limited production. The
Congress has now capped the program at twenty operational
bombers, and the cost of the program at $44.4 billion. While
there is risk in the program, it has been minimized through
extensive laboratory, simulator, airborne test bed, and flight
test activity. The first operational aircraft will be delivered
on December 17, 1993.
The Minuteman Ill/Life Extension Program will replace aging
components on the Minuteman III weapon system. Little
engineering and design work is expected; thus it has a low cost
uncertainty.
119
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
J
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
LINE NO.
INSERT NO.
Question 5
B-2 VS TRIDENT--COST PER SURVIVING WARHEAD
Senator Glenn:
nuclear and conventi
following approximat
for a B-2 weapon. U
B-2 at about S2 bill
amounts to $125 mill
alert rate of about
of the B-2s would su
arriving B-2 warhead
reasonable? (Part 1)
The B-2 has been designed to have both a
onal mission. Please comment on the
e estimate for the cost per arriving warhead
sing estimates for the life-cycle cost of a
ion (for 20 B-2s) with 16 warheads each, this
ion per warhead. If the B-2 has the B-52
one-third, this implies that about one-third
rvive an attack, raising the cost of an
to about $375 million. Is this approach
The Navy has stated that two-thirds of the Tridents are at
sea and survivable, on a day-to-day basis, and about 80 percent
of the Tridents would be survivable under a generated alert.
Using the day-to-day alert figure and the $110 billion total cost
of the Trident program, we get about $50 to $100 million for each
arriving warhead, depending on whether 8 or 4 weapons are placed
on each SLBM. Is this approach reasonable? (Part 2)
From the above results, is it reasonable to conclude that
the B-2 is some four to seven times more expensive for each
arriving warhead? (Part 3)
Dr. Perry: Part 1. The approach is not reasonable.
Bombers are not currently on alert. Any situation that would
change that condition would more than likely mean a full
generation of the B-2 force. Assuming a one-third alert rate is
therefore an inappropriate assumption. If the B-2s are fully
generated, the computation for the cost of an arriving warhead
would be $125 million based on a life cycle cost of about $2
billion per bomber.
Part 2. Again the assumptions are incorrect. To keep a B-2
and a Trident comparison on the same plane, one would have to
assume that the set of events that caused the generation of the
B-2 fleet also changed the posture of the Trident fleet.
Assuming a fully generated Trident fleet of 21, and assuming a
cost of $110 billion and four warheads per missile (2,016
warheads), the cost of an arriving warhead would be $54 million.
Part 3. From the above analysis it would appear that the
B-2 is two to three times more expensive per arriving warhead,
not four to seven times more expensive.
However, such analysis is immaterial at this time. The
correct time to use such a cost-per-arriving warhead analysis is
when one is making decisions on what force structure to buy,
evaluating alternatives, etc. To do so after the forces have
been ordered, delivered, procurement programs truncated, and/or
canceled is not meaningful.
The modernization of the Strategic Triad started in the
early 1980 's and addressed the three legs of the Triad as a
package.
120
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
• SENATE
HEARING DATE
Jun 101993
TRANSCRIPT PAGE NO.
UNE NO.
INSERT NO.
Question 5
B-2 vs. Trident — Cost per Surviving WH (Cont)
The chart below shows what computations of the cost of
'acquired warheads" may have been at that time.
Weapons
per
Total
Estimated
Cost/
Weapon
Quantity
vehic;
e
weapons
cost 1981
warhead
B-1B
90 PAA
16
1440
$20. 5B
$14. 2M
B-2
120 PAA
16
1920
$36. 2B
$18. 8M
Peacekeeper
100
10
1000
$33. 0B
$33. 0M
Trident
21
192
4032
$110. 0B
$27. 3M
(24x8)
From this rebuilt analysis of the cost-per-warhead delivery,
one can see that the B-2 was very competitive when the aircraft
buy was 132 aircraft.
The question is posed in "surviving warhead" terms, and
alludes only to the vehicle being on or off alert as the
determining factor of the survivability. Survivability of the
warhead vehicle is only one element of a very complex issue that
is more appropriately expressed in terms of damage expectancy of
a set of targets.
121
INSERT FOR THE RECORD
Mouse
SENATE
HEARING OATE
Jun 10 1993
APPROPRIATION J COMMITTEE
TRANSCRIPT PAGE NO.
SENATE
ARMED SERVICES COMMrTEE
HOUSE
SENATE
INSERT NO.
OTHER
Government Affairs
Question 6
ICBMs vs SLBMs
Senator Glenn: Do you agree that in all essential
5 ltnltVt "pect3 the Peacekeeper and the Trident D-5 with the
W-88 warheads are essentially equivalent?
*„* JSH'*9??*! Jce Peacekeeper missile with the W-87 warhead
and the Trident D-5 missile with the W-88 warhead share many of
the same warfighting attributes. The yield, accuracy, and range
nard-targ^tmi:rs.eSSentiany e*uivalent< «"«« both effective
fh„ However, a comparison of missile attributes alone ignores
the advantages of the different and independent basing node, of
the Tra;HnK }and"bfsed le°s °f the Triad. Each of the legs of
for J hfM ?? ""i^^^ities to America's strategic deterrent
lea- icSSi ir ^mfSXle 3ubmarines represent the molt survivable
leg, ICBMs are the least expensive to operate and are highly
responsive; and bombers provide recallable, reusable assets. The
elch ?.» / effeCt =reated W combining the unique qualities of
Mot «i?2 h ^KeedSt^e SUra o£ their ^dividual contributions,
anv noiLd°e? ^S T"ad comPii"te the war-planning efforts of
""h° nia adversary, it hedges against the unforseen cata-
strophic failure of a single leg and ensures that the United
deterrenf n°TK° ,\ feliant on anv °ne system for its strategic
Tr^H The qualities provided by all three legs of the
value' withVr-hafy ?"" Feliability, provide adequate deterrent
STAR? II " 6 envisioned for compliance with
122
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
J
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
LINE NO
INSERT NO.
Question 7
MINUTEMAN III MODERNIZATION
Senator Glenn: GAO has concluded that major upgrades of the
Minuteman III are not advisable at this time. Do you agree with
GAO's extrapolation of the DoD estimate of $16 billion by 2010 to
$23 billion by 2020? If the Minuteman III is not upgraded,
approximately when would its reliability be reduced to
seventy-five percent? Would it be prudent to postpone the
modernization of Minuteman III, saving some $23 billion, while we
watch events in the former Soviet Union? Would an unmodernized
Minuteman III be one of the first systems eliminated as part of a
much deeper cuts treaty which included all five nuclear weapons
states?
Dr. Perry: The GAO has quoted a cost of $23 billion to
"modernize" the Minuteman III. The accurate description of the
Minuteman III Life Extension Program is the cost to operate and
maintain the Minuteman III. The Department has documented this
cost as $4.6 billion for acquisition and $11.9 billion for
operations and support, stated in FY 1992 dollars, and covering
the period FY 1992 to FY 2020. The acquisition portion of the
Life Extension Program will accomplish all the required
development and procurement of components that must be replaced
due to age, reliability, or sustainability problems. It also
includes the test costs of a sufficient number of test flights to
verify that the performance of the system is not changed by the
modifications.
The GAO's extrapolation of the Department's estimate of the
cost to operate and maintain the Minuteman III adds $631 million
per year for the period 2010 to 2020, to the $16 billion cost
identified in the Minuteman III Life Extension Program Report to
Congress. The $631 million amount is the annual cost for
operations and support, and is a validated number in that
context. However, because no validated DoD effort has been
undertaken to quantify any additional life extension requirements
for the period 2010 to 2020, we are unable to verify the accuracy
of the GAO estimate.
No validated analytical technique can predict with certainty
when, or if, the Minuteman III reliabilitly would be reduced to
seventy-five percent if the Life Extension Program was postponed.
Postponement of the Minuteman III Life Extension Program can
never save $23 billion. As explained above, this number is
derived from the period 1992 through 2020. It includes
acquisition costs as well as operations and support costs. Even
if all ICBMs were eliminated from the Triad today, savings would
not add up to $23 billion after costs to deposture and store or
eliminate the missiles and facilities were added to the amunt
already spent since 1992.
The most important consideration, however, is not cost.
Retirement of the entire ICBM force would be inconsistent with US
national security requirements in a START environment and would
remove a significant incentive for Russian implementation of
START I and II. Continued deployment of the ICBM force provides
a visible American strategic deterrent force.
123
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Govenmcnt Affairs
SENATE
SENATE
J
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
UNENO.
INSERT NO.
Question 7 (Cont)
MINUTEMAN III MODENRNIZATION (CONT)
Each of the legs of the Triad brings unique qualities to
America's strategic deterrent force: ballistic missile submarines
represent the most survivable leg; ICBMs are the least expensive
to operate and are highly responsive; and bombers provide
recallable, reusable assets. The synergistic effect created by
combining the unique qualities of each leg far exceeds the sum of
their individual contributions. Not only does the Triad
complicate the war-planning efforts of any potential adversary,
it hedges against the unforeseen catastrophic failure of a single
leg and ensures that the United States is not overly reliant on
any one system for its strategic deterrent. The qualities
provided by all three legs of the Triad, survivability and
reliability, provide adequate deterrent value with the force
levels envisioned for compliance with START II. Even under the
most optimistic projections, the START II treaty will not be
fully implemented for almost a decade. Unilateral retiring of
the ICBM force would reduce prematurely U.S. strategic forces to
levels well below the planned post-START II force levels and
completely remove one of the key elements of the strategic Triad.
124
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
J
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PACE NO.
LINE NO.
INSERT NO.
Question 10
B-52 LIFETIME
Senator Glenn: "How many more years can the B-52HS be
expected to last? How many miles have they flown, and what is
their expected lifetime mileage?"
Dr. Perry: The B-52H was designed for 30,000 flying hours.
Currently, there is an average of approximately 13,000 hours per
airframe. Their service life (air worthiness) is projected to
reach beyond the year 2030 based on current tactical operations.
Senator Glenn: "Do you agree with the estimate of $35
billion to maintain 95 B-52s for 30 years? For comparison sake,
what are total costs to go for the B-1B (95 aircraft) and the B-2
(20 aircraft) for the next 30 years?"
Dr. Perry: DoD completed a 20-year estimate for the B-52H
to comply with Section 132 of Public Law 102-190. This Bomber
Comparison Study covered the years FY94 through FY13. All
estimates are expressed in terms of FY92 base year dollars. Our
20 year estimate for the 95 B-52Hs is $17. 7B, which equates to an
average annual cost of $885M. The Air Force update to that 1992
study estimated a cost for 95 B-ls of $20. IB, which equals an
average annual cost of $1B.
Regarding the B-2, there has been no final decision on the
B-2 support concept. DoD continues investigating the optimum mix
of organic and contractor logistics support. Therefore, the
December 1992 Selected Acquisition Report (SAR) contains the best
life cycle cost estimate at this time. The SAR estimate is based
on a nine-year phase-in and a 25-year steady state period of
operations. The life cycle cost estimate for 20 B-2s in FY92
base year dollars averages $542. 0M per year. Over the 34 year
period this equates to $13. 5B. After completing the B-2 logistic
support plan, updated DoD B-2 life cycle cost estimates will be
available.
125
INSERT FOR THE RECORD
SENATE
APPROPRIATIONS COMMITTEE
HEARING DATE
Jun 10 1993
TRANSCRIPT PACE NO
HOUSE
SENATE
ARMED SERVICES COMMITEE
INSERT NO.
HOUSE
SENATE
OTHER
Government Affairs
Question 15
TRIAD COST-EFFECTIVENESS FOR THE FUTURE
Senator Glenn: Is it the position of the Department that in
folce!1"9 °Ur "UClear f°rCeS We aEe seekin* Purely cost effective
e'orhare we constructing a threat-based Triad that is as
cost-effective as possible, given JimWoolsey's "... jungle filled
with a bewildering variety of poisonous snakes....?" rmea
t-h^ ?L £""*; Mthou9h c°st-effectiveness is one criterion
that the Department uses when upgrading strategic nuclear forces
it is only one of many factors considered. ThI DoD intends
through programming decisions and arms control support, that the
Irobu^ H^"69^ nUCiear f°rCeS Pr°vide the United States with
thll *^h d T' Thee und«lying principle continues to be
charac?eri!ti«%iegS °^ thC Triad bringS Uni<^e' complementary
characteristics to our force posture. ICBMs have accuracy,
In thHrVf rellaSi1±tY' and provide a hed9« ^"nst a^roblem
" hae,SLBM force- The SLBMs possess the same attributes, whil^
ICBM for^Pr°TKd%f^edge againSt a P°te"tial problem with the
ICBM force. The SLBMs are also highly survivable when they are
at sea. Heavy bombers can deliver their weapons with high
degrees of accuracy and flexibility, including the feature that
stfhM3? b%r^all?d be£°re reachin3 their targets, adding to the
thf H^fy °f.thiS X!9 °f the Triad- In sum' ^e Triad, trough
the different attributes of the three legs, accomplishes
fdve'rfary6 * ^^ comPlicatin9 attack planning for any future
withinhRussn and encou"9ed fay the progress of democratic reform
lrSh. i\k f neW lndePen<*ent states, we still have many
STA^T trtltie* *rT ?*?f *TP°2 reducti°" negotiated in the *
START treaties are realized. Even after START II limitations are
Tresult "tne'n ?~ V?1 P°SSeSS a "Zable -clear™ 'sen!? £
a result, the Department sees a continuing need for a flexible
Indad,°f K^9^ nuclear forces- I" addition, the capabilities
deler »nH Ulty °J ^ Triad Wil1 Permit the United States to
deter and respond to threats that may emerge in the future The
Nuclear Posture Review, to be conducted by9the DoD? will review
all aspects of our nuclear posture in the post-Cold War world
BOSTON PUBLIC LIBRARY
126 3 9999 05982 898 6
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
V
SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
UNE NO.
INSERT NO.
Question 16
DOD POSITION ON GAO EVALUATION
Senator Cochran: Given that the GAO report was started in
1990, and that several programs considered in the analysis have
since been dropped or dramatically reduced, what is the DoD
position on the GAO evaluation?
Dr. Perry: The DoD position on each finding and
recommendation from the eight GAO reports on the U.S. Strategic
Nuclear Triad is part of the eight departmental responses
officially provided to the GAO and submitted herewith for the
record. The dramatic impact of changing world events and the
fact that, in response to these world events, several strategic
programs were canceled (e.g., Small ICBM, SRAM-II) or reduced
(e.g., B-2, ACM) have been repeatedly addressed in the
Department's responses.
A DoD position that particularly merits further emphasis
concerns the inherent conventional capability of strategic
bombers (i.e., B-52, B-l, B-2) and cruise missiles (i.e., ALCM,
ACM) . GAO evaluations of the various Triad weapon systems have
not adequately addressed the current and/or planned conventional
capabilities of these long-range weapon systems, which take on a
greater importance as the U.S. downsizes force structures,
reduces overseas military presence, and takes on initiatives such
as counterproliferation. The Department will not make
modernization or acquisition decisions for these weapon systems
based solely on their contribution to the Strategic Nuclear
Triad.
127
INSERT FOR THE RECORD
| HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
| SENATE
SENATE
\S" SENATE
HEARING DATE
Jun 10 1993
TRANSCRIPT PAGE NO.
LINE NO.
INSERT NO.
Question 18
TRIAD SUPPORT
Senator Cochran: Cost effectiveness must be a serious
consideration in any acquisition decision, though not to the
exclusion of other considerations, not the least of which is the
question of how to provide the President with the maximum number
of options should there be a need for the use of our strategic
forces. I'd like to put that aside for a moment, though, because
I believe that the GAO report only answers in part what I
consider to be a second-order question. The first-order question
has to do with the viability of our nuclear weapons production
complex: Given that we no longer have the facilities to produce
tritium, I am not aware of any plan to produce tritium in the
future, and the half-life of tritium is 12 years, are we not —
by conscious policy, or worse — proceeding down the road of
unilateral nuclear disarmament, making this discussion of
upgrades to the Triad almost irrelevant?
Dr. Perry: We have no intention of abandoning the Triad nor
placing the efficacy of the US nuclear deterrent at risk by not
ensuring that we have a reliable source of tritium. In
conjunction with the Department of Energy, we are taking steps
to ensure that the nuclear stockpile has an adequate supply of
tritium, not only to replenish it in weapons periodically, but to
ensure we have sufficient quantities to support plausible
contingency scenarios that may place presently unanticipated
demands on the reserve maintained by the Department of Energy.
128
INSERT FOR THE RECORD
HOUSE
APPROPRIATIONS COMMITTEE
HOUSE
ARMED SERVICES COMMITEE
HOUSE
OTHER
Government Affairs
SENATE
SENATE
l/ SENATE
HEARING DATE
Jun 101993
TRANSCRIPT PAGE NO.
LINE NO.
INSERT NO.
Question 19
RESUMPTION OF UNDERGROUND TESTING
Senator Cochran: As you are aware, most of our warheads
were built in the 1960s and '70s, with a small number built in
the 1980s, posing a reliability problem. Once the temporary test
moratorium ends in the near future, will we have the capability
to test? What is the Administration's position on the number of
tests per year? Is this number for a set number of years, or ad
infinitum?
Dr. Perry: Recent and projected retirements of older
warheads will have a significant impact on the age of what we
term the enduring stockpile — those weapons that we plan to
maintain in the post-START environment. With a few exceptions,
these weapons were produced in the 1980s. Each has undergone a
number of developmental and one or more post-production or
stockpile-confidence tests. Based on these tests, coupled with
an active quality assurance and reliability non-nuclear testing
program, Secretary O'Leary has certified that the current stock-
pile is safe, reliable, and secure. It should remain so for a
number of years without the aid and insight derived from further
underground nuclear testing.
In July, President Clinton announced that the U.S. would
continue its moratorium on nuclear testing through at least
September 1994, so long as no other nation conducted a test. On
October 5, despite heavy international pressure not to do so,
China conducted an underground nuclear test. In response to this
test, the White House announced that the President had directed
the DOE to take such actions as are needed to put the U.S. in a
position to conduct nuclear tests next year, provided the
Congressional notification and review conditions are met in the
Spring of 1994. The ultimate decision on whether the U.S. will
resume nuclear testing will be based on fundamental U.S. national
security interests, taking into account all appropriate factors
and considerations. The Administration has begun consultations
with Congress and our allies on these issues. The Clinton
Administration remains committed to the goal of completing a
Comprehensive Test Ban by 1996. To this end, multilateral
negotiations will begin in the Geneva-based Conference on
Disarmament in January.
o
ISBN 0-16-043971-X
9 780160
43971
90000
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Open Library
American Libraries
TV News
Understanding 9/11